NATURE . 
725 

May 26, 1923] 
Societies and Academies. 
Lonpon. 
Royal Society, May 17—A. E. H. Tutton: (1) 
A universal interferometer. The essential feature is 
a travelling microscope driven by a specially con- 
structed fine screw along a true V-and-plane guiding 
bed; one of the two glass reflecting surfaces is 
carried rigidly with it, and the amount of its motion 
is measured directly in monochromatic interference 
bands. There is an autocollimating telescope with 
micrometer eyepiece, a vacuum tube on the elbow 
tube, a constant deviation prism for the selection 
of the monochromatic radiation to be used, and 
large truly worked glass interference discs. The 
telescope is mounted to the right, and the driving 
wheel to the left, and the 30-inch long V-and-plane 
bed, in its rigid carrying plinth-bed, is supported 
on pillars at the Airy positions for no flexure, the 
whole being mounted on a heavy rectangular base. 
The large middle space on the latter is available 
for a large circular work-taple with every possible 
requirement of adjustment for supporting the object 
or its manipulating apparatus. (2) A wave-length 
torsometer, and its use with the universal inter- 
ferometer. This is a refinement of the Voigt instru- 
ment for determining the torsion constants of small 
bodies. It is essentially a miniature lathe-bed, 
carrying two similar but mutually reversed wheel- 
and-chuck fittings, the chucks for gripping the object 
bar ends, and the wheels, which move solidly with 
the chucks, for delivering the force-couple at one end 
and holding the object firmly at the other, the two 
ends being interchangeable. The power band passes 
round the lower half of one pulley-wheel and thence 
over a larger pulley-wheel on a standard. The end 
depending from in front of the latter terminates in 
a loop of the cord-band, into which the hook carried 
by the weight can be allowed slowly to fall, until 
the whole weight is acting in twisting the object 
bar. The torsometer is supported on the work-table 
of the universal interferometer. It is rigidly clamped 
with the two aluminium radials carried by the object 
bar near its two ends, in contact near their upper 
terminations with the blunt knife-edge ends of the 
two sliders. The Grayson-ruling signal is centred 
under the microscope waen the radial under observa- 
tion is just in complete contact with the slider. On 
delivering the weight and effecting the twist, the 
slide and signal move, and the movement is followed 
by driving the microscope by the big wheel of the 
interferometer until the signal mark is again centred, 
the number of interference bands effecting their 
transit being counted—-L. N. G. Filon and F. C. 
Harris: On the diphasic nature of glass as shown 
by photo-elastic observations. A block of flint glass 
was heated to about 400° C., when it showed some 
signs of softening ; it was then allowed to cool under 
longitudinal pressure. On removing the pressure it 
was found to have become permanently doubly- 
refracting. The residual stress which should produce 
the observed amount of double-refraction does not 
balance according to the laws of statics. It is deduced 
that a ‘‘crypto-stress’’ exists, which does not 
manifest itself optically. This leads to the conclusion 
that the glass is not homogeneous, but behaves as 
a mixture of two components or phases.—C. E. 
Inglis: Stress distribution in a rectangular plate 
having two opposing edges sheared in opposite 
directions. Imagine a thin rectangular plate bounded 
by two horizontal lines AB CD and two vertical 
lines AD BC. The two horizontal edges, while 
remaining straight and unchanged in length and in 
NO. 2795, VOL. 111] 
distance from one another, are displaced longitudinally 
in opposite directions, the vertical edges being kept 
free from applied stress. The plate being thin, 
the distribution of stress consequent on this deforma- 
tion is regarded as two-dimensional and the stress 
_components are obtained through solutions of y* V =o. 
Along the horizontal centre line the stress starts 
from zero at the free edge, increases rapidly, and, 
for a plate in which the length is considerable com- 
pared with the depth, the stress soon assumes a 
constant value; but before doing so, it overshoots 
this value, and the curve of stress distribution in 
consequence develops humps near the free vertical 
edges. If the length—breadth ratio of the plate is 
2 to 1 these humps combine to give a flat-topped 
curve. If the plate is square the coincidence of the 
humps makes the curve approximate to a parabola. 
—T. H. Havelock: Studies in wave resistance : 
influence of the form of the water-plane section of 
the ship. In these calculations the ship is represented 
by a vertical post of infinite depth the horizontal 
section of which is similar to the water-plane section 
of a ship. The level lines of the model are varied, 
while the displacement is kept constant. In this 
manner a comparative study is made of such problems 
in ship resistance as the effect of finer lines and 
greater beam and of the difference between straight 
and hollow lines—W. M. H. Greaves: On a certain 
family of periodic solutions of differential equations, 
with an application to the triode oscillator. There 
is, under certain conditions, a discontinuous family 
of periodic solutions of the equations d/dt =né, 
dy/dt =\(«) +n, where \ (x) is a function of x only, 
€ and 7 are functions of x and y, periodic in y with 
period 27, and expressible as Fourier Series in sines 
and cosines of multiples of y, the coefficients being 
functions of ¥, not involving ¢ explicitly, and » is 
a constant parameter. An application is made to 
the equation of Appleton and Van der Pol for the 
triode oscillator the equation of which can be reduced 
to a particular case of the above equations. 
Geological Society, April 18.—Prof. A. C. Seward, 
‘president, and, afterwards, Dr. H. H. Thomas, vice- 
president, in the chair—J. F. N. Green: The 
structure of the Bowmore-Portaskaig district of 
Islay. Quartzite is defined as containing a limit of 
ro per cent. of felspar, more highly felspathic rocks 
being termed “ arkose.’’ On this definition the Islay 
upper quartzite is throughout true quartzite; but 
the so-called lower quartzite is not quartzitic, being 
composed of arkoses and greywacke-slates identical 
with the matrix of the Portaskaig conglomerate ; 
they have been grouped together as the Portaskaig 
beds. Thus the dolomitic group intervenes between 
the Portaskaig beds and the Islay quartzite. The 
supposedly Torridonian Bowmore sandstone consists 
of arkoses and flags. The flags, which, owing to 
isoclinal folding, have apparently an enormous thick- 
ness, are identical in minute detail with certain 
siliceo-argillaceous flags that always occur in_ the 
dolomitic group next to the Islay quartzite. They 
are termed the Bowmore flags. Thus there is no 
change of facies at the supposed thrust, and its 
presumed line of outcrop shows that the rocks are 
folded up without disruption, except for some shear- 
ing on the reversed limbs of overfolds. The Loch 
Skerrols thrust is non-existent. The Bowmore flags 
are perfectly conformable to the white edge (a well- 
marked horizon) of the Islay quartzite ; the dolomitic 
flags associated with them are partly or wholly 
cut out in places by the Portaskaig beds. The latter 
are probably younger than the dolomitic group. 
On Beannan Dubh the rocks lie in isoclinal folds 
with low dip, by which the Portaskaig conglomerate 
