98 MONTHLS WEATHER REVIEW. MARCH, 1899
20 .............. 21. ............ 22 .............. gj ............. 24 ............ I .............. ZR ............ 27 .............. 28 ..............
Means.. .......
OBSERVATIONS AT S P. 1. EASTERN ETANDARD TIME, (5217 P. M. LOCAL.) ____________.___ -
Tempera-) 1 I Gpper clouds. I I Lower clouds.
1 ture.
79 79 8n.5 R I
78 Ro i 9
79 3
$8.2
10
3 8 10
._
0
78 09
72
7a
i a 52
i 4
74
71 73 i 3
i 3 6.5
w 72
?d i 3
i 3 i3 i 3
51 i 4 72
;.a 75
i 8 75
.... -
ne.
se. ne. se.
...............
..............
-
ne. ne. ne. ne. ne.
se. ne. se. se. se .
se.
Re.
ne. ne. se. ne. ne. se. se.
ne. ne.
88.
W.
e.
St?.
ST.
E.
Sr?.
....
-.
1::::::::l i I.::::::::: ........ 0 .......... ....... ......... ....................... ........................ ........ 0 ......... ....... 0 ......... ........ 0 .......... ........ 0 .........
10 ne.
1 e.
.... ................
.............. I ..........
ck. 8 I e. ........I ................
........ I ..... I .. ...... ..........
........ ..........
.. .....
I ~
I .
2 24
1
2 1 0 0.5
0
s 0.6
1 1
8-4
2
1 1 2 1 1 2
1 2
0
1 3 a
I
, -
...... ....... ...... ....... ks. ka. ....... ....... ....... k.*
H k. k.
ak. d. k. f . k. rk.
nk. ak.
k. a.k.,k k.
f. k.
k.
.......
.......
.......
.......
......
..... I ........ I ...... I .......... I I I I ....... ~
:umnli on Ometepe.
The rainfall occurred as follows : 3d, sprinkle at 3 a. m.;
9th, rain a t 3:15 and 9 a. m. ; 19th, thunderstorm from 7 to
8 p. m.; 17th, sprinkle. 0.02 inch a t 1 n. m., frequent showers
reported at Tortuga, about 50 miles southeast of Rivas on the
southwest shore of Lake Nicaragua; Mth, sprinkle at 5:15
p. m.; 19th, sprinkle, 0.10 p. ni.; %et, sprinkle at 1 p. ni.
The barometric range for the month was 0.16. The lowest
occurred on the 21st and the highest on the 14th. Cool
waveB occurred on the 9th and 14th. On the 8th calm and smoky with a light air from the southwest : a shower occurred
5 miles to the northward, and a sprinkle a t Rivas; 9th,
wind backed to northeast a t 10 a. m.; 15th, phenomenal
clouds from the south and southwest.
MEXICAN CLIMATOLOGICAL DATA.
Through the kind cooperation of Sefior Mariano BBrcena, Di-
rector, and Sefior Jose Zendejas, vice-director, of the Central
Meteorologico-Magnetic Observatory, the monthly sunimaries
of Mexican data are now communicated in manuscript, in ad-
vance of their publication in the Boletin Mensiml; an abstract
translated into English measures is here giveu in continua-
tion of the similar tables published in the MONTHLY WEATHER
REV~EW siuce 1896. The barometric means have not been
reduced to standard gravity, but this correction will be given
a t some future date when the pressures are published on onr
Chart IV. Narkan. &ti5 .for MarcJi. 1Y99, -
I
‘6
stations. i 2
I =
!a
.. .- ...
--
i z
0 3,. 91.8
93.2
8;. 3 88.5
m. 4 64.4
92.1 88.5 loo. 4 82.9
I. 7
-_
OF.
85.2
99.8 99. a 43.6 88.1
8u. 8 5 i .
41. ‘ 44.6
53.8 I
4J.a
1
OF.
74.7 0.2. e 65.8 64.2
81.9 65. P
ill. 7
p. 8
18.3 BB. i 89.4
$3 !&I
88 2i w 46
55 59
7G 89
85
1 Inch.(
...... sw. ...... ...... ......
8W. e. 0w.w.nw sw. sw. W. sw. ssw. n.
W.
wsw.
-_
WEATHER FORECASTING IN HONGKONG.
By W. DOBERCX, Dlrwtor of the Hongkong Observatory (dated February 17,1899).
In the law of storms in the eastern seas it is esplained that
all the phenomena connected with typhoons are natural con-
sequences of the barometric gradients, and that the steepness
of these cause enormous rainfalls, and that these tend to in-
crease the gradients till the rainfall ceases for lack of water
vapor when the center of the typhoon enters dry land. These
phenomena are not qualitatively different from those expe-
rienced in colder climates. Althongh the climates feel so
extreniely different, there is scarcely sufficient difference in
temperature to cause any substantial difference in the laws
governing tlie weather. This is most apparent when the es-
treme dilTerences in temperature are expressed on the abso-
lute scale beginning’with absolute zero.
In the northeast nionsoon the wind blows practically always
froni tlie northeast, east, v n east-southeast, as pressure is rela-
tively lowest to the south. In midwinter the lowest pressure
lies to tlie south of the equator, and in spring and autumn i t
lies to the north of the equator, a trough-shaped depression
lying between tlie northeast, a d southwest winds. On the
contrary, in the southwest monsoon there is no southwest
wind in Hongkong unless there happens to be a deprekion
to the north of the observer. A permanent depression inland
in northern China or Siberia does not exist.
During the northeast m‘onsoon, when the center of an ant. - cyclone moving along eastward between preceding and fol-
lowing cyclones, passes canparatively close to Hongkong the
weather clears there. The latitude of the centers of the anti-
cyclones is generally about 35O, and perhaps never as low as
37O. The time when the northeast wind is strongest is not
when the center is just north of or nearest to Hongkong. but
occurs usually when the center is past, because the high pres-
sure spreads to the south and southeast, so that pressure con-
tinues rising along the south coast of China after the center
is past.
When during the northeast monsoon a low pressure ad-
vances across north China and Korea it seldom causes south-
west wind in Hongkong, but only calms or very light winds.
A t the same time southweat winds are frequently reported
from Saigon and the southern Phillipines, apparently against
the gradient. This is caused by local shallow low pressures over the land. which becomes iiiteiisely heated, owing to the
absence of the usual northeast monsoon and owing to the
clear sky and hot sunshine. Such southwest breezes must
have a diurnal period like land uncl sea breezes, and they do
not blow a t sea except very near land.
Northers in Hongkong are just like northers in Texas.
They occur with falling temperature after very hot days in
winter and spring. In case of high barometric areas over
north China, Korea, and Japan sometimes a V-shaped depres-
sion with isobars open toward the Routh is formed near For-
mosa. Such a depression develops into a cyclone moving
toward Japan.
While the weather in Hongkong in winter depends upon
the latitude in which tlie cyclones and anticyclones are cross-
ing to the northward, it depends in siimnier upon the latitude
of the troughs.
Mr. A. G. Figg, who officiates as weather forecaster in
Hongkong, states that there appears to be a general agree-
ment in recent years between droughts in India and droughts
in Hongkong.
Before a period of foggy weather sets in we note an upper
current from south or southwest above the east wind. Then
fog occurs along the coast, which is cooler than the sea, .with
light (usually east) mind. With west wind the coast is not
so cool, and therefore fog is not so likely to occur as with
east wind or calm.
MARCH, 1899. MONTHLY WEATHER REVIEW. 99
Thunderstorms occur when gradients disappear with change
of gradient, for instance, occasionally, before a typhoon ;
when the weather is hot; and especially, when the wind is northwest and it is very hot inland in China. Mr. Figg has
sometimes noted jumps in the barometer readings before ae
well as during thunderstorms. He states that when they
approach from the landside there is very little rain, while
if they come up from the seaside there is great rain.
SELENIUM AND ITS USE FOR THE MEASUREMENT OF
SUNSHINE.
By N. ERNEST DORSET. Ph.D., of Johns Hopklns University (dated April 17,1899).
Owing to frequent inquiries as to tlie suitability of some
form of the seleuiiim cell as a continiiws and exact sunshine
recorder, i t has been deenied advisable to publish in this RE-
VIEW a short account of what is known in regard to the selen-
ium cell, especially with respect to this use.
The fact that the resistance of selenium is changed by the
action of light was first announced by Willoughby Smith in
1873. He wished to use selenium bars with platinum wire
electrodes melted into their ends as high resistances to be
used in connection with submarine cables. His assistant
found that the resistances of these bars were very inconstant,
and this variability was found to be due to the varying
illumination of the bars. It was found that the decrease of
resistance noticed when the bars were illuminated was due
to the visible radiation, and appeared to be instantaneous.
Smith suggested that this effect might possibly be explained
by the fact that selenium conducts electricity only when in
the crystalline condition, and that light favors crystallization.
Immediately upon the publication of Smith's paper Lieu-
tenant Sale, and also the Earl of Rosse, repeated and veri-
fied Smith's observation. The latter suggested that this
property of selenium might be used as a means of measuring
the intensity of light, as he found that the decrease of resis-
tance is almost proportional to the quare root of the inten-
sity of the illumination.
I n 1875 Werner Siemens went over this work again and
then undertook the study of the effect of the physical state
of the selenium upon its seusitiveness to light. He found that by protracted heating of amorphous selenium at 210" G.,
or by cooling melted selenium to thiR temperature (at which,
with a longer duration of it, the selenium passes into a
coarsely granular, crystalline state), he obtained a modifica-
tion of crystalline selenium which possesses and retains a
considerably greater conductivity than otherwise. It is also
far more sensitive, and the decrease of its resistance due to
its exposure to light appeared to be constant. He constructed
the first of the so-called selenium cells, which he describes
as follows : " By fusing into coarsely granular selenium two
flat spirals of wire a t the distance of about one millimeter
from each other, I produced an extraordinarily sensitive pho-
tometer." Obscure heat rays were without eRect upon this
cell, while diffused daylight doubled its conductivity, and
direct sunlight increased its conductivity, a t times, tenfold.
Prof. W. G. Adanis and R. E. Day now undertook investi-
gations on this subject and in 1877 they published in the
Philosophical Transactions a long and exhaustive article 011
the subject. Besides obtaining results similar to those
already described, they found that the resistance of the sele-
nium cell generally decreased with an increase of the current
through it, and depended upon the direction of the current.
They also found that the cells became polarized on the pas-
sage of a current ; that the change of resistance clue to illu-
mination depended upon the end illuminated ; that when no
current was passing through the cell the action of light on
the cell could give rise to a current through it, hut in regard
to this latter property they say :
It appears that three pieces of the same length, made from the same rod, and annealed to ether may, owing to some slight difference in molecular condition, %e very different as to their relative sensitiveness to the action of light.
It was also observed that a slight heating produces a great
increase in the resietance, which also changes very greatly
with the time; they appear to anneal slowly.
I n 1878 Sabine studied the resistance of the fused-in elec-
trodes and found that it was very great and depended upon
the direction, strength, and duration of the current passing
through the cell.
In 1884 C. E. Fritts descrihed a new form of the selenium
cell which is much more sensitive than any of these others.
He melts a thin layer of seleniuni upon a metal plate with
which it will form a chemical combination a t least sufficient
to cause the selenium to adhere to i t and make good electrical
connection, tlie other surface of the selenium is covered with
a transparent conductor, generally gold leaf, through which
the current is passed into the selenium. Like other kinds of
selenium cells the resistance of these depends upon the
strength and direction nf the current and the temperature
and age of the cell.
In 1888 Uljanin investigated this subject and explained
the observed phenomena much in the same way as Smith
first suggested. He assumed that the annealed selenium con-
sists of various allotropic forms of selenium, some of which
are conductors and some are not, and the action of light i s
supposed to favor the change from one form to the other.
He gives a good rheum6 of the work up to t,his time. I n a series of articles published in the Philosophical Ma- gazine, from 1881 to 1595, Bidwell points out that all the
properties of the selenium cells which have been so far dis-
covered suggest that the concluction through these cells is electrolytic.in character. The selenium is probably a non-
conductor, and the current is carried by metallic selenides contained i n the selenium. In support of this he finds that
selenium which has never touched metals but has been an-
uealed in glass has a much higher resistance than that an-
nealed in the ordinary way ; and, furthermore, its resistance
is decreased by the addition of metallic selenides, so that i t behaves like ordinary selenium. He also found that cer-
tain speciniens of selenium which were entirely insensitive
to light, were rendered very sensitive by the addition of selen-
ides. And, finally, he succeeded i n constructing a cell coni-
posed of sulphur and silver ~iilphide which behaved exactly
like the selenium cell. His last article, published in the
Proc. Phys. SOC. 13, 1891, pp. 552-579, and Phil. Mag. (5 ) 40,
1896, pp. 233-256, contains the hest and most recent r&suniB
of the entire subject.
From this we see that while selenium cells may be used for
very rough determinations of the intensity of illumination,
they are eminently unsuitable for any exact photometric work. Owing to the fact that the resistance varies with the
strength and duration of the current, and with the tempera-
ture, and with the entire past history of the selenium, each
cell would have to be carefully studied in order to obtain the
coefficients of these various factors, and after this was done
these coefficiente would be correct for but a very short time,
on account of the unknown and variable change of the resist-
ance aud electromotive force of the selenium cell with its age.
The best articles on this subject are as follows: Willoughby
Smith, Nature, 5, 1573, pp. 303 and 361 ; Am. Jour. Science,
(3) 5. 1873, p. 301. Lieutenant Sale, P. R. S., 2:, 1873, p.
283; Pogg. Ann., 150,p. 333; Phil.Mag. (4) 47,1814,~. 216;
Earl of Ross, Phil. Mag. (4 ) 47,1874, p. 161 ; Weruer Siemens,
Monntsberichten der k h . preuss. Akad. d. Wissenschaften zu
Berlin, 1S75, p. ?SO; Phil. Mag. (4 ) 50, p. 416; Pogg. Ann.,
149, p. 140; W. 5. Adam8 and R. E. Day, Phil. Trans., 167, 1877, pp. 313449; Proc. Roy. Soc.. vols. 83,2425; Sabine,
Phil. Mag. (5) 5, 1878, p. 401 ; Bell, Proc. A. A. A. S., 99, 1580;