more properly ended in a sphere; the lower end
was connected with a pair of gold-leaf strips, as
in the earliest form of electroscope. This appa-
ratus was improved by attaching a slow-burning
match or flame at the upper end of the rod. The
best of the early forms of apparatus was that of
Dellman, in which a brass ball is electrified by
induction and afterwards its exact condition is
determined by applying an electrometer. The
modern method devised by Sir William Thom-
son (Lord Kelvin), about ISfiO, consists of a
reservoir of water, supported on insulating glass
pedestals; a small insulated tube leads from the
reservoir to some point in the open air, at a dis-
tance from the building, and the water flowing
through this tube forms a stream that is broken
into drops within a few inches of the end of the
tube. This constitutes the water-dropping col-
lector. In freezing weather the water must be
replaced by a vessel of oil, with wick and flame.
The most modern apparatus actingbyinduction is
that devised by Jlr. Morrill, of the United States
Weather Bureau. If the electric potential of the
tube and the water at the point where the stream
breaks into drops differs from that of the adjacent
air, the drops will carry away the positive or
negative excess, and soon bring the collector to
the same potential as that of the air. The electric
condition is observed by measuring, not sjjarks
or currents, but the difference of potential that
tends to cause such currents. For brevity, the
word 'potential' is used to signify difference
of potential. The difference of potential between
the air and the earth is measured by connecting
the ground and the collector to the opposite poles
of some form of electrometer, and for this pur-
pose the quadrant electrometer of Kelvin has
been most widely used, since it was first intro-
duced at the Kew Observatory in 1861. A
modification of all this apparatus has been used
in France since 1875, knowii as Mascart's sys-
tem, and this apparatus has also been used by
the United States Weather Bureau. In 1884
that bureau inaugurated a systematic inves-
tigation of this subject by starting observations
and studies under the guidance of Professors
Rowland, at Johns Hopkins, and Trowbridge, at
Harvard I'niversity. This work was subsequent-
ly placed under the control of Prof. T. C. Alen-
denhall, whose report was published in the Mem-
oirs of tile Nationnl Academy of Sciences, Vol. V.
(1889). The specific object of this investigation
was to determine whether there was any apparent
connection between the electrical condition of the
atmosphere and the development and progress of
storms, so that such observations could be util-
ized in the improvement of storm and weather
predictions, or in drawing lines of equal electric
potential on the Daily Weather Map. It was,
however, demonstrated that the potential de-
pends so much upon local conditions that observa-
tions made in the same neighborhood gave entire-
ly different results, and that it was necessary to
learn how to interpret the local record and its
oscillations before combining it with records from
another station. As it did not seem likely that
results of any value in weather forecasting could
be secured at present, the further prosecution of
the subject was discontinued. The principal work
that has been done since 1887 has been that of
Elster and Geitel in Germany, Maseart and Cha-
veau in France, and J. J. Thomson and C. T. E.
Wilson in England.
As regards the general phenomena of atmos- pheric electricity, observations throughout the globe harmonize in showing that, in genei'al, in calm, clear weather a difference of potential ex- ists such that the air is positive and the earth negative ; this difference is larger during east and northeast vinds in the Northern Ilemis- phere, and it oscillates violently during thunder- storms; there are also two daily maxima and two minima; the potential difference is much larger in winter than in summer, except during thunder- storms; the difference increases greatly during falling snow and during strong winds. Lord Kel- vin suggests that there may be cloudless masses of air having different charges of electricity float- ing above and producing the observed changes in potential as they pass by the observer. As regards the origin of atmospheric electric- ity, a great number of hypotheses have been ad- vanced and discussed ; most of the older ones have been shown to be unsatisfactory, but the newest suggestions are still undergoing discus- sion, and our knowledge of the whole subject re- mains in an elementary condition. Perhaps the nearest approach to a plausible explanation is found in the work of J. J. Thomson and C. T. R. Wilson, who have been able to show that when aqueous vapor begins to condense from the air it settles by preference on the particles of dust that have a negative electrical charge and that a con- siderable degree of supersaturation is required to make it condense upon those that are positively electrified. It follows that the first formed par- ticles of fog or cloud are negative and are larger and, therefore, heavier ; and that they, settling to the earth as rain, give the ground and the lower portion of the atmosphere a negative charge rel- ative to the upper strata of air. This explana- tion, is, perhaps, better than another which has had many adherents, namely, that the earth, be- ing negatively electrified, induces a positive charge in the air. The action of ultra violet radiations in discharging a negatively electrified body is the same as though they them- selves were positively electrified ; and this has led to the hypothesis that the radiation from the sun may give a positive charge to the upper laj-ers of the atmosphere. The hypo- theses that atmospheric electricity is due to the evaporation of water at the earth's surface, or to the friction of the wind, or to thermo-electric cur- rents, or to the discharge of great volumes of steam through volcanic vents and geysers, must all be given up as insufficient. If it should be found that we must return to the original theory of Peltier that the earth has a negative charge, and that the atmosphere becomes positive by in- duction, then the most plausible suggestion — but as yet undemonstrated — is that of Clerk Max- well, namely, that the tidal strains within the earth's crust and the slipping that cause faults and earthquakes give rise to a development of piezo-electricity that has gradually accumulated until the present condition has been attained. This hypothesis connects terrestrial electricity and terrestrial magnetism together, and may be considered as a new example of the transforma- tion of energy, since it is equivalent to the daily conversion of a small fraction of the force of gravitation into electricity.
For a general summary of our knowledge of atmospheric electricity at the present time, see Hann, Lehrbuch der Meteorologie (Leipzig,
