JUNE, 1928 MONTHLY WEATHER REVIEW 247
tion were greatly increased. In the early seventies the
necessity of linking the coastal stretches with the com- mercial telegraph systems of the country was an outstand- ing problem for the Army Signal Service. A beginning
was made in New Jersey where the first unit of what was for many years known as the “sea coast” line was con-
structed between Seaville and Pecks Beach, N. J., a stretch of but 10 miles. Immediate steps then were taken
to construct a line along the beach of the New Jersey coast from Sandy Hook to Cape May Point and later to
extend that line to Smithville, N. C. The line was fin- ished in about a year and functioned successfully for many
years; the section from Cape Henry to Hatteras, N. C.,
now transferred to the Coast Guard was a part of the
original construction. The line was operated directly from the signal office in Washington, D. C., and had for
its purpose the display of storm warnings in the interest of coastwise as well as of across-seas traffic. Another factor in its use was the succor of vessels in danger of foundering or in distress. Communication between ship
and shore was had by means of the international signal
flags and by visual signalling in the rare cases when a
Signal Service man boarded a vessel in distress and wished
to communicate with shore.
Military telegraph lines in the interior, as from time to
time, authorized by Congress were for the most part con- structed by troops detailed for that purpose. Among the
first, if not the very first line so constructed, was one
joining San Diego, Calif., by way of Fort Yuma and Maricopa Wells, Ariz., with Prescott and Tucson. This
construction was followed by a survey and preliminary work looking to the building of lines connecting military
posts in Texas, of which a t that time there were 10 or 12.
Some of them were along the Texas-Mexico boundary, one was in the Texas Panhandle and others were more or
less distant.from the advance posts of settlement. To
connect these points and other places along the border
required a greater outlay of time and labor than had hitherto been espended in any single State.
Concurrently with the activity in Tesas, building new
lines and extending those already built both in Arizona
and New Mexico also in the far Northwest was being
carried on.
‘The longest stretch of line when single units were
joined together was that extending from San Diego, Calif., to Denison, Tex., as an eastern terminal. The present writer, when stationed at the last-named point in 1879,
well remembers making a number of attempts to work with San Diego, but without success due, no doubt, to
the poor insulation of the line in places. Ordinarily the attempt to work long stretches of the military lines as a
single circuit was not made. The Texas lines of a total length of more than 1,500 miles were operated as a single circuit three times daily in the collection of meteorological
reports from the stations on those lines. At other times they were operated as two or three separate circuits.
Several causes were responsible for the gradually dwin- dling mileage of the military telegraph lines from its peak of 5,114 miles in 1882 to 1,025 miles in 1891. These causes, named in the order of their importance were, (1)
appeals to Congress for increased appropriations for their
maintenance were only partially realized, (2) the custom of detailing enlisted men from the ReguJar Army as opera-
tors and linesmen failed in 1883, (3) the abandonment of military posts naturally resulted in the sale or dismantling
of the line if local interests were not sufficiently great to
warrant its maintenance as a private venture.
In 1883, the year aft,er the peak was reached 2,450
miles of line were sold or abandoned and eight years later
when the meterological activities of t’he Army Signal Service were transferred to the newly created Weather Bureau in the Department of Agriculture but 1,025 miles
of the original 5,114 remained. Only those sections that were vital to the meteorological service were continued in
use by the Weather Bureau. The new construction by that bureau amounted to a total of 270 miles of land lines
and submarine cables; that mileage plus the 629 miles
inherited from the Signal Corps makes a total of 899 miles all of which has now been disposed of either by sale or
transfer as noted in the beginning of this article.
EFFECT OF CLOUDS ON THE SURFACE TEMPERATURE
By W. J. HUMPHREYS &-S/. b-76 : SC/. 3-#24
Obviously the radiation emitted by and from any por-
tion, large or small, of the surface of the earth tends to come into equilibrium with the radiation simultaneously
absorbed by the same surface. Clearly, too, this es-
change, though. generally equal only twice in the course of a day and %&t, would, 09 the average, balance Per-
fectb (neglecting the minute SUPPlY of heat from the in-
terior of the earth) if there were no conduction to and from the atmosphere, nor vertical or horizontal motion-
convection and advection-of the air or the oceans, nor
evaporation or condensation. But all these things do
occur and they greatly disturb the radiation balance.
=Owever, be
clear or cloudy, and therefore may be disregarded in corn-
the surface temperature.
bclosing, area is a function of the material and nxdmn- ical condition, rough or smooth, of the surface (reentrant angles producing a closer approach to the full radiation)
its temperature (directly proportional, nearly, to the fourth
power of the absolute temperature), and to the barometric pressure, varying directly as the square of the refractive
index of the adjacent air. This latter effect is negligible, since the refractive index in question differs very little from unity. The rate of this radiation does not, how-
ever, depend a t all on the state of the sky, such as clear
or cloudy.
On the other hand, the rate of absorption by the given
radiating surface does depend, and very greatly, on the state of the sky owing to the consequent large variations
in the amount of incident radiation that might be ab-
sorbed. It also, like emission, is a function of the mate- rial and condition of the surface and of the barometric
pressure.
In general, except as prevented by winds, convection
and evaporation, t,he temperature of the surface tends
sorption. Furthermore, the greater the rate of incident radiation the greater, in substantially the same ratio, the
rate of absorption and the higher the surface temperature.
Let-
s= the net radiation received (incoming less out-
going) per unit horizontal area dunng a clear
night.
are roughly the same whether the
a fist approximation to the effect of clouds on rapidly to become such that emission is equal to ab-
The rate Of emission per unit projected, Or
6 4 7 4 6 2 6 2
JUNE, 1929 248 MONTHLY WEATHER REVIEW
&''=the net radiation similarly received from sun
C= the net radiation received per similar area from
C'=the net radiation received per like area from a
Evidently, then, if, as assumed, we may neglect every- thing but radiation and absorption, and consider the co-
efficient of absorption to be the same whether the sky be
clear or clouded, the effect of a sheet of clouds is to
lower the surface temperature, to leave it unchanged, or to raise it, according as S+S' is greater than, equal to,
or less than C+ C'. From observations by Kimball,' we know that through-
out the night, and for the latitude of Washington, D. C., the net outgoing radiation is, when the sky is clear,
0.15 calories, roughly, per minute per horizontal square centimeter, and 0.05 calories when the sky is covered
with clouds. Furthermore, from observations, S = 5 C, roughly.
Finally, let X be the number of minutes from sunset to sunrise.
At Washington, D. C., the total radiation received er
square centimeter horizontal surface during a clear f a y is: in June, 732 calories, .and in December, 241 calories.
That is, numerically, in June S'=732, C'=146, and in
December, S' =241, C'=48. In June, X =600, while
in December, X=880. Hence in June, S = -90,
and during a clear day.
a low cloud canopy by night.
low cloud covering by day.
1 MONTHLY WEATHER REvmw, 46, p, 57,1918.
8 Marvin and Khball, Journal of the Franklin Institute, 202, p. 302, 1926.
C= - 30, and in December, S = - 132, C= -44. Thus, a t Washington, in June, S + S' = 642 and C+ C' = 116.
Therefore a cloud canopy, day and night, at Wash- ington, D. C., would lower the temperature in June.
In December, S+ S' = 109, C-?- C' =4. Hence in Decem- ber also a continuous cloud canopy would lower the tem-
perature a t this locality. At a somewhat higher latitude,
however, probably around 50') the cloud canopy would not change the temperature at this season, while at a
still higher latitude it would raise the temperature.
As stated above, evaporation and condensation, and
the circulation of air and wat>er are very effective as
distributors of heat. Therefore the boundary along which a cloud canopy would have no effect on the surface tem-
perature is distorted irregularly in time and place by all
these phenomena, as well as more or less unifornily shifted
with the course of the seasons. Only continuous and direct observntions can give us full information as to the
places and times of the warming and the cooling of the surface of the earth-the places of net gain and net loss
of heat by all processes combined. However, it does
seem practically certain that a continuous cloud canopy over the entire earth would materially lower its average
temperature. I t would raise the temperature of polar and high latitude (beyond latitude 50°, roughly) winters,
and lower the temperature at nearly all other times and
places.
This qualitative result is, of course, unsatisfactory, but
that appears to be all we can obtain a t present with
assurance.
WINTER OF 1928-29 IN EUROPE 5-5/. 5 g ( 41
By W. R.
[Weather Bureau. Wn
The past winter has been one of the most severe that Europe has experienced since the inception of systematic
meteorological observations. I n Berlin, for example, the
mean winter temperature was the lowest of record, with but one exception, 1829-30. There was but one dny
from December 8, 1928, to March 7, 1929, when the teni-
perature was not below freezing. December was about 2O F. below normal, January 6' F. below, and February
20' F. below. Since 1851 there have been but six winters
with all three months below normal. February, 1929, was the coldest since meteorological observations were
begun a t Berlin in 1720; the temperature of - 13' F. observed on the morning of February 11, is the lowest of
record. Lowest temperatures of record were also ob-
served at Hamburg (-6' F.), Hannover (-13' F.), Frankfort on the Main (- 7' F.), Frankfort on the Oder
(-24' F.), Dresden (-18' F.), Leipzig (-17' F.), Breslau (-26' F.) Munich (-25" F.), and Vienna
(- 15' F.), the latter being the lowest since observations were begun in 1775. During January pressure was above normal over west-
ern Europe and greatly above in the region of Iceland,
Isafjord being 0.72 inch above, while Horta was 0.33 inch
below. One of tjhe most unusual features in January, 1929, was the fact that pressure averaged higher over Iceland than over Horta.
In February also the presswe distribution was very
abnormal. Iceland and Horta returned to nornial, or
slightly above, but over Scandinavia the departures were
as much asf0.60 inch, while departures in southern
Europe were negative.
STEVEN0
shington, June, 1928J
During the first few days of January a LOW of con-
siderable intensity was central over the Mediterranean,
which was accompanied by heavy rains and resulted in
the worst flood on the Tiber since February, 1915. Cold weather and heavy snows occurred during the first half of the month quite generally over Europe as far south as the
Riviera. In central Europe the snows were so heavy that
railway and telegraph comniunications were broken in several places, the ice on the Elbe above Hamburg was so
thick that the river could be crossed on foot, skating was permitted on the lakes in the Bois de Boulogne in Paris
on the 17th and 18th for the first time since 1917.
The most severe period lasted from approximately
January 21 to February 21. For about a week previous
to the beginning of this period a HIGH of great intensity
had been building up over Siberia in the Provinces of Irkutsk and Yakutsk, and began gradually spreadin
pressure extended from Japan to western Europe with a
crest of 31.39 inches at Bratski-Ostrog in the Province of Irkutsk, which had advanced to western Russia, Perm,
31.16 inches, by the 24th. Pressure fell over southern and central Europe, a LOW of considerable intensity develop-
ing over the Mediterranean by the 25th, which moved eastward and on the 31st was central over Limasol,
Cyprus, 29.59 inches. This LOW was attended by heavy snows as far south as the Riviera and severe cold and
violent storms in Jugoslavia. In the meanfime pressure
remained high in Russia and Siberia, Leningrad, 31.08
inches; Chita, Trans-Baikal Province, 31.30 inchee. By the morning of February 4 the Leningrad HIGH had
to the westward. On the morning of January 21st, hig %