524 
NATUPE 
[JuLy 16, 1914 
sent out horizontally from any ‘point on’ the earth’s 
surface round the earth parallel to its surface are con- 
sidered. Pure diffraction is insufficient to account for 
all the phenomena of long-distance wireless telegraphy, 
but some action of the atmosphere which tends to 
curve the radiation round the earth has to be postu- 
lated. The theory of ionic refraction, based on the 
theoretical conclusion that in ionised air the velocity 
of long electric waves is increased, has been put for- 
ward. The atmosphere decreases in density as we 
rise, and this alone produces a decrease of refractive 
index and an increase in velocity. Formule are de- 
duced expressing the variation of density with heights 
taking into account the known temperature variation 
with increase of height. At a height of 100 km. the 
terrestrial atmosphere must consist substantially of 
hydrogen and helium. An expression is obtained for 
the radius of curvature at any point of a ray of light 
sent out horizontally from the earth’s surface. This 
radius at the starting point is given by p=y,(98Aq,°), 
where w, and q, are the refractive index and density 
at the surface, and A is the Gladstone and Dale con- 
stant for the gas which forms the atmosphere. For 
air p is four times the earth’s radius, for 
hydrogen 136 times, and for krypton equal 
to the earth’s radius. If the terrestrial atmo- 
sphere consisted wholly of krypton a_ ray 
sent out horizontally would be refracted round the 
earth, and wireless telegraphy to the Antipodes would 
be possible. For the same atmospheric density and 
constant A this circular refraction would result if the 
earth were twice its present diameter. The sugges- 
tion is made that perhaps neon and krypton are manu- 
factured at great atmospheric heights by electric dis- 
charges occurring in the rarefied hydrogen atmosphere. 
Also that by their ease of ionisation they contribute 
to produce the ionised layer demanded by the theories 
of Heaviside and Eccles to account for the actual 
achievements of long-distance wireless telegraphy. 
Our earth is perhaps unique in being the only planet 
on which long-distance radio-telegraphy is possible. 
—G. Dobson; Atmospheric electricity observations 
made at Kew Observatory. Observations were made 
(1) using the standard Wilson instrument on a stand 
according to the usual practice, and (2) using an 
experimental apparatus level with the ground, which 
was assumed to give correct results. A comparison 
was made of the electric conductivity of the air as 
measured by Mr. Wilson’s apparatus and that de- 
signed by Prof. Ebert.—T. Barratt : Thermal and elec- 
trical conductivities of some of the rarer metals and 
alloys. A new method of the ‘stationary tempera- 
ture’’ type is employed for measuring the thermal 
conductivities of some of the rarer metals, including 
tantalum, molybdenum, rhodium, iridium, and tung- 
sten, at air temperatures and at 100° C.—F. Mercer : 
Some investigations on the arc as a generator of 
hgh-frequency oscillations. Experiments on the 
copper-carbon arc when used as a generator of high- 
frequency oscillations. The first experiments deal 
with the effect of varying the arc length, and also the 
arc current, on the magnitude and frequency of the 
shunt current. The effect on frequency arises from 
a change in the resistance of the arc. The second 
refers to the effect on the shunt. current of altering 
the ratio of inductance to capacity. 
Paris, 
Academy of Sciences, July 6.—M. P. Appell in the 
chair.—Arnaud de Gramont: General observations on 
the ultimate lines of elements from. various sources 
of light. It is pointed out that the strongest lines 
in the spectrum of a simple body, the ‘‘ Hauptlinien ”’ 
NO. 2333, VOL. 93] 
of the German physicists, are not identical, the ulti- 
mate lines persisting in the condensed spark, and the 
work of Hartley and Moss is criticised from this point 
of view. Arranged in decreasing order of temperature 
the sources of light used were ‘the condensed spark 
with self-induction, condensed spark without self- 
induction, non-condensed spark, electric arc, oxy- 
acetylene blowpipe, oxygen-coal gas flame. — Experi- 
ments were carried out on forty elements, and a 
general summary of the results is given.—M. de 
Forcrand : The thermochemical study of some hydrates 
of manganese sulphate. The values obtained for the 
hydrates with 2, 3, and 4 H.O are not in accord with 
Thomsen’s data for the same salts. There would 
appear to be two isomers of the anhydrous sulphate.— 
P. Chofardet: Observations of the new comet 19g14c 
(Neujmin) made at the Observatory of Besancon. 
Position given for July 4. The comet appeared as a 
round nebulosity, about 15” diameter, with a slight 
central condensation. About 12:5 magnitude.—G. 
Beauvais: The definition of time given by a clock. 
A study of the clock installed in the cellars of the 
Paris Observatory, by means of Abraham’s photo- 
graphic chronograph. It was found that a double 
second might easily be 0-008 sec. too long or too short, 
with occasional rare deviations amounting to 0-02 sec. 
The effect of this on the comparison of two pendulums 
by the method of coincidences and upon the definition 
of time is discussed. Georges J. Remoundos: Series 
of functions and the singularities of differential 
equations.—Th, De Donder and O. De Ketelaere: The 
electromagnetic field of Maxwell-Lorentz and _ the 
gravitation field of Einstein.—Gustave le Bon: The 
principle of relativity and intra-atomic energy.—Léon 
Brillouin; The calorific conductivity and viscosity of 
monatomic liquids.—C. de Watteville : A new method of 
studying spark spectra. It is known from the work 
of Hemsalech that when a spark passes between two 
conductors the initial spark is followed by the produc- 
tion of metallic vapour, and the latter remains 
luminous for an appreciable time. A new form of 
apparatus is described which permits of the separation 
of the luminous effects of the spark and the metallic 
vapour.—G, Brafas: The microradiograph. <A descrip- 
tion (with diagram) of a new self-recording Morse 
apparatus for radio-telegraphic signals. With this 
apparatus installed at Madrid records of messages sent 
from Paris, Poldhu, and Norddeich have been regis- 
tered.—H. Kamerlingh Onnes: The persistence of 
electric currents without electromotive force in super- 
conductors. From a study of the resistance of metals 
at low temperatures attainable with liquid helium it 
was concluded that the resistance of mercury would 
be measurable at 4:25°, but would become negligible 
at 2°. This conclusion has been verified experiment- 
ally, but with the unexpected result that the resistance 
disappears suddenly, for mercury at 4-:19°. In a mer- 
cury thread at 1-7°, current can be passed with a 
density of 1000 amperes per sq. mm. without a 
measurable difference of potential (limit of accuracy 
0-03 x 10° volt) at the extremities, and without develop- 
ing heat. (See article in Nature, July 9, p. 
481).—H. Abraham, A. Dufour, and G. Ferrié: A 
method of direct measurement of the time of pro- 
pagation of the waves of wireless telegraphy on the 
surface of the globe. The chronographic method 
ultilised permits of the absolute measurement of a 
time interval with a precision of o-o0001 sec. The 
velocity of propagation found for the Hertzian waves 
between Paris and Washington was 296,000 km. per 
sec., slightly less than the velocity of light.—M. 
Abonnenc ; The influence of tellurium on the sensibility 
of selenium to light. Carefully purified selenium was 
ae 
