434 
quarter, except in the case of a thunder-storm, when 
the record was set at fifteen minutes. In the 
hottest days of July no test at all could be ob- 
tained and none at night in either month. This 
would seem to indicate that the ultra-violet rays 
of sunlight have a great deal to do with the forma- 
tion of ozone. 
In the bracing days of late November the 
weather was cold and snappy. A test resulted in 
less than twenty-five minutes on November 13. 
During the following three weeks several tests were 
secured in thirty or forty minutes. At this time 
the silent electric discharge between earth and 
clouds must have been quite instrumental in form- 
ing the ozone. The greater stability of ozone in 
cold weather allowed it to accumulate. 
Probably after a cyclonic disturbance of the 
upper air we on the surface get a supply of ozone 
manufactured in the upper laboratories. On the 
occasion of some of the above tests air was drawn 
over crystals of chromic acid to eliminate the 
effect of hydrogen peroxide. Furthermore, deli- 
eate reagents for hydrogen peroxide and nitrites 
failed to yield a test. The inference is that all 
the above tests were due to ozone alone. 
JAMES M. BELL and MELyILLE L. BuckuEy: The 
Solubility of Bromine and Iodine in Aqueous 
Solutions of Alkali Bromides and Iodides. 
C. W. Foutk: A Modified Burette Calibrating 
Pipette. 
W. M. BLancHarD: A Simple Lecture Apparatus 
for Illustrating the Relative Ionizations of Salts. 
ArTHUR B. LAMB and JOHN W. MarpDEn: The 
Quantitative Determination of Perchlorates. 
The determination of perchlorates by fusion has, 
in spite of its great simplicity, been generally 
abandoned in favor of some method of reduction 
in solution. This has been due to the fact that 
during ignition or fusion there have been small 
but apparently unavoidable mechanical losses, pre- 
sumably of potassium chloride. The authors have 
shown that by merely carrying on the fusion in a 
test tube, suitably plugged with asbestos wool, 
instead of in a crucible, these losses could be pre- 
vented, and by precipitation of the chloride as 
silver chloride a very simple but accurate deter- 
mination of the perchlorate obtained. 
G. N. LEwis and F. G. Keyes: The Potential of 
the Potassium Electrode. 
G. N. Lewis and PavL FarraGHER: Potentials be- 
tween Liquids. 
J. E. Smpet: The Molecular and Intermolecular 
Energy. 
SCIENCE 
[N.S. Vou. XXXV. No. 898 
In this paper the author attempts to show that, 
““PaV’’ being practically equal to ‘‘RxT’’ for per- 
manent gases, no notable portion of the kinetic 
energy can be present therein as rotary or inter- 
molecular motion, since neither of these motions 
could contribute to the pressure and volume energy, 
of which a full equivalent appears for every 
amount of heat or kinetic energy imparted to a 
gas, and that therefore all kinetic or heat energy 
of gases must be almost exclusively due to the 
translatory motions of the molecules, while other 
internal molecular motions must be confined to 
chemical, radiant and other energies. Confirma- 
tory of this the author also finds that the trans- 
latory molecular velocity in gases calculated on 
above basis, after different independent methods, 
is uniformly almost exactly double that of the 
velocity of sound in the respective gases, thereby 
indicating another mode of propagation of sound 
in addition to that by aerial resonance, in which 
new mode of propagation the molecular vibrations 
act as vehicle for the sound impulses, which new 
mode of propagation would fully explain all 
acoustic phenomena, including also such (audi- 
bility of very feeble sounds for miles, ete.) not 
explainable by aerial resonance only. 
A. A. Noyes: A Proposed System of Notation of 
Physico-chemical Quantities. 
JOHN JOHNSTON: A New Form of Mechanical 
Vacuum Pump. 
This pump—known as the May Nelson Rotary 
Pump—consists essentially of two plates of metal, 
one fixed, the other movable; no valve, packing or 
fluid is required, though a little lubrication is 
desirable for mechanical reasons. A somewhat de- 
fective pump of this type has given a vacuum of 
0.01 mm. when connected to the gauge by rubber 
tubing. It has exhausted 6 liters to 0.1 mm. in 
2 minutes. After pouring in 1 ¢.c. of water, the 
pressure was reduced to 0.5 mm. in 4 minutes, 
without drawing any air through the pump. The 
exact mode of working is difficult to show without 
a diagram, but the principle may perhaps be made 
sufficiently clear by what follows: 
In each of the metal plates are cut a series of 
concentric grooves, which are so situated that the 
projections on each plate will lie in the grooves on 
the other plate. Through the center of the fixed 
plate passes a shaft, on which is an eccentric 
carrying the movable plate. Rotation of the shaft 
thus produces an eccentric motion—but no rota- 
tion—of the movable plate; as a result there is, 
