333 MONTHLY WEATHER REVIEW. MAY, 1903
companied by a simultaneous increase in the absolute amount
of the charge; at an altitude of 3 kilometers we have a charge
of more than 4 electrostatic units per cubic meter. For ex-
ample on the basis of the Elater-Geitel determinations of the
electrical charges in the atmospheric precipitation, V. Con-
rad (Wiener Berichte, 111, Abth. IIa. p. 342, 1902.) has com-
puted that the amount of electricity in 1 gram of the water
of a cumulus cloud amounts to 1/36 of lo4 of a coulomb.
Within a dense cloud in which the vision could penetrate to a
distance of only 18 meters, there was according to Conrad’s
measurement 5 grams of water per cu1)ic meter, consequently
an electric charge of about l/7 of lo4 of a coulomb per grain
of water. Now if the above-mentioned value of 4 electro-
static units, or 4/3 of lo”= 4/30 of lo-$ of a co111oml) of nega-
tive electricity be assumed as the charge per cubic meter, then
even this amount of electricity would sul-fice to explain qiian-
I n general the process of condensation brings clown only a
small fraction of the electrons present. Suppose now that n e
consider that only the negative electrons take part in the pre-
cipitation, then these are weighted clown by coatings of water
and sink down as rain, but according to our measurements
about an equal quantity of positive electricity per cubic meter
remains behind in the cloud. Nom as Conrad has already
shown, if, for instance, we suppose a cumulus cloud of spheri-
cal form of only 1 kilometer radius to rest with its center 3
kilometers from the earth’s surface, then i t will by its own in-
ternal charge cause a decrease of potential at the earth‘s sur-
face of about 11,000 volts per meter of vertical distance. Now,
such values have been actually observed in thunderstorms at
the earth’s surface. If we reflect that for such a graclient a
point in the air of 500 meters above the earth would sliow n,
difference of tension of 5,600,000 volts with respect to the
earth, then we find ourselves here brought to consider tensions
such as we see relieved by the mightiest electric, process of the
atmosphere. i. e., the thunderstorm. As early as 18S7 Liiis’ had
calculated the immense electrical forces called into being when
the charges assumed by him to exist in a cloud, were separated
by great stretches of space, ancl showed that sources of energy
were here revealed to us, which were more than sufficient to
explain the most violent phenomena of thunderstorms. The
theory of electrons now gives us, as we hare shown, a sur-
prisingly simple explanation 6f these charges, and our elec-
tron traps deliver to us catches whvse iuagnitucles arc quite
sufficient to explain the phenomena i n a quantitatire way.
And now finally the last problem, the one wliicli offered alto-
gether insurmountable dificulties to all the old theories. be-
gins to gradually become resolvable from the standpoint of
the new theory, viz, the problem of the permanent charge of
the earth and the existence above it of a field of electrical
tension, or the so-called ‘( fine weather electricity.”
It was clear even to the earlier observer8 that the earth’s s ~r -
face always possessed an electric charge relative to the atmos-
phere, even in typical fine weather, when there was no trace
of thunderstorm conditions within a considerable radius. At
such times the earth’s mass showed itself negatively charged
as compared with the surrounding air; only during cloucly,
rainy weather inclining to th un ilers torm forma tion. W O U ~J the
sign of the earth’s charge occasinally reverse, but eveu then
only for short periods. To explain this electric charge proper
to the earth itself the most divergent theories have been sng-
gested, but none of them have been proved satisfactory. The properties of the electrons furnish a wholly new point of view
from which the problem appears surprisingly sinlple. The positive and negative electrons are to be distinguished from
one another wherever they occur by the different velocities at
See Pellat tranhlatrd hy A. G. MoAdie, dme~ican Meteorological Jour-
nal September, 1885, Vol. 11, pp. 215-331, and Park Morrill at ~J J . 438-
445 of the same volume.-E,r>.
titatively the observed electricity of precipit a t’ 1on.
~~
which they travel. Under the impulse of a given electrical
force the negative electrons are more easily set in motion and
they travel much faster than the positive electrons, which seem
to be loaded with a greater quantity of inert matter. On the
other hand both positive and negative electrons seem to be
charged with the same quantities of electricity, which are dis-
tinguishable from each other only by their opposite signs.
Now i f such an electrical particle pass near a conducting sur-
face, such as the earth’s surface or that of some conductor in
electric contact with the same, then the passing particle induces
in the conducting surface a superficial charge of the opposite
sign, which attracts the particle to it. This attracting force,
which is directly proportional to the square of the charge and
inversely proportional to the square of the distance of the
particle from the conducting surface, influences both species
of electrons in the same way, but the negative are able to re-
spoud to the electrical forces more easily and quickly than the
positive. Thus, during a unit of time and with equal charges
of positive and negative electrons in the air, a larger num-
ber of negative thau of positive electrons will always reach
the conducting surfaces ancl give up their charges to them.
On mountains, tree tops, and similar places this process is of
subordinate importance, since on these projections the charges
of the negative earth repel the negative electrons and collect,
BH we have seen above, a preponderating number of positive
electrons. There are, however, many spots on the earth’s sur-
face where its own charge is without effect in reference to the
particles in the air and where, therefore, the inflow of negative
electricity can proceed undisturbed. As Elster and Geitel
have showii these places are all concavities, particularly those
occurring under the leafy roof of the earth’s vegetation, which
are of the greatest estent, but also the cavities formed by
caves, chasms, and fi8sures. I n the latter cases the projecting
portions and points form a very perfect electrostatic protec-
tion against the electrical field of the earth, which otherwise
woulcl hinder the wandering of electrioity into the charged
ground. We have evidences that the vegetation in particular
plays a very important part in the atmospheric electric pro-
cesses, ancl that the process above explained is quantitatively
sufficient to renew the electric charge of the earth in the man-
ner just described. It is certain that such a renewal of the
earth’s charge must occur, since the air is not a perfect electric
insulator, and the conductivity due to the wandering of the
electrons causes a perpetual tendency to equalization of the
earth’s charge ancl of the gradient of tension in the atmosphere.
There is still much to be said on the subject of the relation
of this latter gradient to the conductivity of the air and the
charge of electrons, and there is already a rather ertensive
collection of observations at hand which opens a series of new
and interesting perspectives. A further consideration of this
subject on this occasion would lead us too far; but we may
rejoice in the fact that in the theory of electrons the processes
of atmospheric electricity have acquired a point of view that
promises to contribute very much to the solution of problems,
some of wliicli are centuries old, and that incites US to the most
zealous pursuit of further studies in this much contested field
of research.
ABNORMAL VARIATIONS IN INSOLATION.
R) Mi ti. 11. KIND41 L, h.i*tant Editnr. d a t d April 15, 1W3.
I n the Comptes Rendus. Paris, March 16, 1903. Volume
C’SSSVI. pp. 713-715, Monsieur Henri Dufour announced
that hi^ observations with a Ckova actinometer at Lausanne,
Switzerlnncl, indicated a diminution in the ainount of solar
radiation received a t the surface of the earth at that place
(latitude 46” 3 8 ’ ) in January, February, and March of the
present year, as compared with the average of corresponding
months for previous years. This is shown in the following
table:
MAY, 1903.
Month. 1897- 1902.
January. ..... 0.79 February .... 0.86
March. ....... 0.89
MONTHLY WEATHER REVIEW.
1903. Difference.
~_____~
0.6R 0.11
0. 71 0. 15
0. 70 0. 19
233
1902.
1903.
January ........
~~
Average solar radiation at noon at Laueanne, S-land, in gram-enloria per equare centimeter per mini&.
~~
0 0
36.2 .............................. 31.2 ..............................
3s. 2 ..............................
89.3 ............................. .................... 0.9% Ji. R
~~~ ~ ~
-~
M. Dufour is inclined to attribute this deficiency to the pres-
ence of large quantities of volcanic dust in the air as the re-
sult of the eruptions of last year in the West Inclies.
It is evident that tlie solar radiation of M. Dufour is the
radiation from the sun, as received by us on the earth, after
it has been diminished by the very appreciable losses due to
absorption and other atmospheric influences. This insolation,
as actually measured by physicists, is expressed in gritiu-
calories per square centimeter per minute. It has regular
diurnal and annual variations but the abnormal variations are
those that we are now considering.
Observations of insolation were imde by nie for the United
States Weather Bureau with an h ~g s t r h i n electrical coinpen-
sation pyrheliometer, from November 10, 1902, to March %;,
of about 2200 feet and a latitucle of alJOUt Y G O 36'. There are
no previous observations a t tliese points with which to coiupare
results, but it mas noted at the time and was the occasion vf
comment, that tlie measurements did not increase after Dr-
cember as much as had been expected. The folloniiig are the
monthly averages for the dates of observation, at iiooii, in
gram-calories per centimeter per minute:
1903, at Asheville ancl Black Mountain, N. C., a t an elev n t' 1on
not sufice for the observations made in this country when
considered by themselves, but i f the insolation was deficient
over most of the Northern Hemisphere, and continued to
be deficient for a period of several months, then some more
general explanation must be sought for. If volcanic dust is
the cause, no doubt it will manifest itself in other ways, as, for
instance, by causing brilliant aft,er glows following the usual
sunset colors. Observations of insolation and of sunsets for
t,he coming months should therefore have special interest for
meteorologists.
~
The following monthly means of insolation observed with the
Bngstriim apparatus a t Washiiigton, D. C., a t noon on clear
clays during April, May, and June, 1593, are added as this note
goes to press:
Solnr rcrtlicrtioii, i l l grciiri-cnlorirs per nqiscrre cmfintefer per miiiicte.
I--I
1!1113.
~
I April _. ._ ._. . I
____
HAILSTORMS IN PORT0 RICO.
BJ bf1. lv. H .i L l \ \NI>ER, O h e ~\e i \~cnthaT Bllrr.au, dlnlral Alwil 70, 1903
Hailstorms itre so rare in Porto Rico that the impression
seems to be quite general even among Porto Ricans that they
never occur. This is a mistake as wah recently demonstrated.
The change of senson from winter to summer occurs about
the middle of April :mil is, as a rule zery iuarked, being char-
acterized by uiiusnnlly n ariu, sultry clays, frequent thunder-
storins, the setting in of the trades, ancl in a iueasure the Le-
ginning of tlie so-called rainy season. This sew was no ex-
ception unless i t be in the unusud strength with which the
trades have set in. The recortls show that from the 11th to
the l i t l i thunderstorms were quite geileral over the island.
The only important. because unusual. feature of the season
worthy of special iiieiitiou was the occurrence of a heavy pre-
cipitation of hail on the 12th instant in the vicinity of Caguas.
Thinking a report of this iuiglit l ~e of interest, efiort has been
macle to secure as full ani1 izccurate inforiuntiou relative thereto
as poshible. Two intelligent gentlemen, one an American and
the other R Porto Rican, who were eyewitnesses of the event
have been interviewed, and their reports are fully reliable and
confirm other information obtained froiii other sources.
The forenoon of Sunclay, April 12, 1903, was warin ani1 sul-
try, very faroralde for tlie developmelit of thunderstorrus. It
appears that the storm now under consideration liacl its begin-
ning about 2 1). iu. in the neighborhod of Agnas Buenas,
moved eastward along and down the 1 alley of tht. Bairo River,
acrnss the Loizn, and up the alley of the Gurabo. The storm
was accoiupanied by some lightning ant1 thuncler and very
violent w i d s , rendered more violent ant1 destructive, no doubt,
by the peculiar topogralhy along the storin's track. Some
small huts were overturned and considerable claiuage done to
the uncut tobacco along the valley of the Bairo River.
Hail was first observed ttt ilguas Buenas where, as reported
1)y RIr. Bowser, the fall mas light, lasting :tlwut ten or twelve
minutes. but hrther down the river the fall wits so heavy that
the river 1 ~~1 was b L n-liite as snow," so thick were the hail-
stones. The track of the storiii qqwars to have been just
north of C'tlguns, although hail fell there for almit fifteen inin-
uten acscor&iig to Dr. Lngoviiio, who mas in tlie city at the
time. The 1)recipitation of hail coutinuecl its far as Gurabo,
but how much farther is not known.
A s to the tiize aiic1 form of the ldbtunes, there >are Heveral