L110 

integer in the formula, or, as they may be called, the 
spectra of various orders, differ amongst themselves in 
a surprising way. The intensities in the case of the 
most important planes in Iceland spar have recently 
been determined and give very interesting results. In 
the case of two pairs of planes the spacing is the 
same but the arrangement of atoms is different. This 
gives an opportunity of comparing the effect of 
arrangement apart from spacing, and it appears that 
the intensity of the reflection in any order is propor- 
tional to the weight of the planes which contribute 
to that order. Again, there are three calcite planes 
for which the arrangement of the atoms is exactly 
the same, but they differ in their spacings. The rela- 
tive intensities follow a rule which has already been 
stated, viz., that the intensity in a reflection at an 
angle @ is inversely proportional to sin *6, other things 
being the same. Rules of this kind are needed if the 
method is to be used in the examination of more 
complicated crystals. The physical meanings that 
may be attached to these rules are of considerable 
interest. 
Physical Society, February 26.—Dr. A. Russell, vice- 
president, in_ the chair._Dr. C. Chree: Magnetic 
“character” figures, Antarctic and international. The 
paper makes use of magnetic ‘‘character” figures 
“9” (quiet day), “1” (moderately disturbed day), 
“5% (highly disturbed day) to investigate whether 
the incidence of disturbance at the base station of the 
Scott Antarctic Expedition, 1911-12, did or did not 
accord with the incidence of disturbance in temperate 
latitudes; also whether the ‘“57-day period”’ could be 
recognised in the Antarctic data. The incidence of 
disturbance in the Antarctic was found to agree closely 
with that shown by the international lists, in spite of 
the fact that the disturbances in the Antarctic were 
much larger and more persistent than at any of the 
stations co-operating in the international scheme. The 
‘97-day period” was clearly visible in the Antarctic 
records both in summer and winter, being as well 
developed there as elsewhere.—Dr. P. E. Shaw: The 
electrification of surfaces as affected by heat. The 
paper deals with anomalous electrical behaviour of 
various substances when subjected to heat. For 
example, a glass rod rubbed with silk is normally 
left positively electrified, but if the rod be passed 
through a bunsen flame, or heated in an electric 
furnace, and then allowed to cool it will be found 
on again rubbing with the silk that the glass becomes 
negatively electrified. Similar results were obtained 
with a number of materials, and various experiments 
are described which aim at determining the cause of 
the phenomenon. These seem to show that it is not 
due to the formation or removal of layers of any sub- 
stance, solid or gaseous, but is probably due to surface 
strains in the material—Prof. J. W. Nicholson : 
Electromagnetic inertia and atomic weight. The 
paper contains a mathematical deduction of a simple 
formula for the combined mass of two electrical 
charges when in proximity to each other. This mass 
is not the sum of their individual masses when far 
apait, if it be supposed that all mass of positive 
electricity, like that of electrons, is of electromagnetic 
origin. | Applications are made of the formula to 
questions of atomic constitution and of radio-activity. 
A discussion is given of the evidence leading to the 
conclusion that the nuclei or cores of positive elec- 
tricity in atoms are complex structures of electrons 
and even smaller positive nuclei. On this basis, 
emission of an a particle by an atom does not decrease 
its atomic mass by 4, a correction being necessary 
for the ‘‘mutual mass”’ of the a particle and the rest 
NO. 2369, VOL. 95] 
NATURE 

[MarcH 25, 1915 

\ 
of the core. Estimates of the magnitude of this 
correction, in the case of radium and thorium passing 
into lead by the emission of particles, are given. 
From the value given by Soddy for the atomic weight 
of thorite lead we can deduce the average distance 
apart of the components in a radium nucleus. It is 
of the same order as the radius of an electron. Sug- 
gestions of further interesting applications of the 
precise formula for mutual mass are also contained 
in the paper 
March 12.—Dr. A. Russell, vice-president, in the 
chair._C. C. Paterson and B. P. Dudding : The estima- 
tion of high temperatures by the method of colour 
identity. Preliminary experiments are described in which 
the temperature of incandescent substances is esti- 
mated with a very fair accuracy by matching their 
colour with that of incandescent filament lamps work- 
ing at appropriate efficiencies. These have previously 
been calibrated by comparison with a “black body ” 
over a large temperature range. The comparisons are 
made in the field of a Lummer Brodhun photometer, 
and the method is shown to give the correct result 
for the melting point of platinum.—C. C. Paterson 
and B. P. Dudding: The unit of candle-power in 
white light. The paper describes the methods adopted 
at the National Physical Laboratory for minimising 
the difficulty of photometric comparison of white lights 
of different hue. A series of six sets of electric sub- 
standards are described varying in the tint of the light 
radiated from that of the pentane lamp to that of a 
tungsten vacuum lamp operating at 1-5 watts per 
candle. The absolute value of the unit of candle- 
power has been re-determined, as have also the cor- 
rections for humidity and barometric changes, while 
the probable existence of a temperature correction is 
discussed.—G. L. Addenbrooke: The relative losses in 
dielectrics in equivalent electric fields, steady and alter- 
nating (R.M.S.). After references to former work, 
especially on surface leakage, tables are given show- 
ing an intimate connection between the losses in 
steady and alternating fields, and that one can be pre- 
dicted from the other to a first approximation. Ex- 
ceptions are mentioned, and it is shown that there is 
heterogeneity in these cases. Curves of the losses 
from 1 to 4 seconds to 40~ are given for specimen 
dielectrics. Above 8 to 12~ both for “good” and 
‘poor’ dielectrics, these become rising straight lines. 
The salient features lie below about 16 ~. The 
formula a+bn applies to all dielectrics for frequen- 
cies above about 8~, but not below. The a constant 
is always larger than, and bears only a very indefinite 
relation to, the steady voltage loss. Liquid dielectrics 
behave similarly to solid with certain differences, par- 
ticularly that for a given resistance the ratio of the 
steady to the alternating loss is less than with solids. 
There is a great difference in the comparative ranges 
of the losses. For the dielectrics tabulated these vary 
in a steady field from 1 to 1,300,000, but the corre- 
sponding alternating losses vary only from 1 to 1,100. 
Linnean Society, March 4.—Prof. E. B. Poulton, 
president, in the chair.—a. W. Hill: The germina- 
tion of the cucurbitaceous genus Marah. The genus 
Marah includes some eleven species practically con- 
fined to the Pacific watershed of the western States of 
North America and the islands of Lower California. 
The genus is distinguished from Echinocystis and 
other genera with similar floral characters by its 
enormous tuberous root, associated with which is the 
peculiar mode of germination of the seeds. The 
petioles are fused to form a tube, and on germination 
this tube grows out, carrying plumule and radicle 
some distance into the ground, the cotyledons remain- 
