Marcu 7, 1907] 
NATURE 
449 
in 1904-5 differ radically in showing no southerly com- 
ponent, the Otaria was sent again to the south and west | 
of the region which had been explored in the preceding 
summer. Mr. Rotch states that the upper anti-trade is 
shown both by the balloons and the drift of the clouds 
between 3000 metres and 4ooo metres, and that the classic 
observations of the return trade on the Peal of Teneriffe 
indicate a general phenomenon, and agree with those 
obtained over the open ocean by the recent expedition. 
Prof. Hergesell's remarks upon this subject were referred 
to in NATURE of December 27, 1906 (p. 211). 
Meteorological Observations on the Summit of the 
Tsukubasan, Japan.—The establishment of this first-order 
observatory, and the determination of the force of gravity | 
and exact geographical position, are due to the interest 
taken in physical science by H.I.H. Prince Yamashina. 
The observatory is situated on the most westerly peak of 
the mountain, in lat. 36° 13/ 21” N., long. 140° 5! 47! E., 
about forty miles north-east of Tokio, at an altitude of 
2852 feet; it commands the view of the surrounding dis- 
trict for many miles to the north and west, while to the 
south and east it has an open view of the wide expanse | 
of the Pacific Ocean. Its position is therefore extremely 
The Mete:rolcegical Observatory on the Tsukuba-an, 
favourable for studying the conditions of the atmosphere 
at that height. As connecting links, intermediate stations 
have been established near the little village of Tsukuba, at 
an altitude of 787 feet, and at the base of the south-west 
of the mountain, 98 feet above sea-level. The illustration 
represents the peak observatory, which is constructed of 
wood and zinc, the main objects being durability and use- 
fulness, without any attempt at ornamentation. On the 
roof are seen the rain-gauges, lightning conductor, and 
wind-vane; close to the main building, on the north-east, 
stands a steel tower carrying another lightning conductor, 
anemometers for recording both horizontal and vertical 
movements of the wind, and a sunshine recorder, while 
the thermometer screen is seen to the south-west of the 
building. It goes without saying that the instruments are 
of the best make, although the sunshine recorder is of 
the photographic (Jordan) type, not the burning (Campbell- 
Stokes) pattern. The latter instrument alone is now used 
at the stations of the British Meteorological Office, as 
giving strictly comparable results. The observing staff 
consists of a director and five assistants; at the time of 
the publication of the first report, for the year 1902, the 
observatory and subsidiary stations were under the super- 
vision of Mr. Okada, adjunct of the Central Meteor- 
ological Office at Tokio, the control of the observers and 
other details being undertaken by Mr. J. Sato, chief 
NO. 1949. VOL. 75] 
observer at the peak station. The computation of the 
mean and extreme values for 1902 from hourly readings 
for all three stations, and tables showing the ranges and 
wind frequency and velocity, are carefully prepared, but 
no general textual summary of results is given. At the 
base station only the rainfall observations are complete 
for the year; the total fall was 63-72 inches, and at the 
peak station 62-82 inches. The. absolute ranges of the 
barometer at the summit and intermediate stations were 
practically the same, being 1-87 inches and 1-78 inches 
respectively. The mean annual temperatures at these two 
stations were 48°.2 F. and 55°-2 F., and the absolute ranges 
39°-2 and 37°-4 respectively. The resulting wind direction 
at the summit, computed from the records of a Robinson's 
anemometer, was N. 82° E., resultant velocity 0.39 metre 
per second (0-87 mile per hour); the mean hourly velocity, 
irrespective of direction, was approximately 17-2 miles. 
Meteorology of India.—The Meteorological Department 
of India has issued a memorandum on the weather con- 
ditions during October and November, 1906, with a fore- 
cast of the rainfall in northern India and of the snowfall 
on the neighbouring mountain areas during the cold 
weather of 1906-7. Dr. Walker states that on the average 
of the whole country there was a defect 
of per cent. in the rainfall of 
October and of 20 per cent. in 
November. The temperature conditions 
were determined by the distribution of 
rainfall; in the latter month the 
weather was unusually warm over prac- 
tically the whole of the country, 
and especially in the North-Western 
Provinces. From information avail- 
able, the snowfall also appears to 
have been less than usual. Among 
other factors affecting the  cold- 
weather season, the director points out 
(1) that the active state of the sun 
during the past year is an element that 
should be taken into account; the 
number ef sun-spots observed in 1906 
is in moderate excess, which fact, if 
taken alone, suggests that a severe 
winter is rather more likely than a mild 
one. (2) That the mean of the de- 
partures of November rainfall at Zanzi- 
bar and Seychelles is —1-8 inches, 
which, taken by itself, suggests that 
the approaching cold-weather  precipi- 
tation may be somewhat lighter than 
usual. All things considered, the final 
conclusion is that there is no reason 
for expecting any large departure from 
normal conditions. 
Meteorological Observations in Cape 
Colony.—Yhe report of the Meteorological Commission 
for the year 1905 shows that a large amount of useful 
7 ; circum- 
22 
work is being carried out in rather adverse 
stances. The sum received from the Parliamentary 
Grant for the year did not exceed 862l.; the supply 
of instruments and reduction of anemometrical _and 
other observations have consequently been curtailed, 
while no general inspection of stations has been made 
since 1901. The results are published for a_ large 
number of ordinary meteorological and rainfall stations, 
some of which belong to adjacent territories outside the 
boundaries of Cape Colony; the report also contains a 
useful monthly chronicle of the weather by Mr. C. M. 
Stewart (secretary), and special tables of the maximum 
daily rainfall at various stations. The mean rainfall for 
the year, deduced from all the stations, was 23-77 inches, 
occurring on sixty-five days; the amount was only about 
0-1 per cent. below the average for 1885-1894, and was 
an increase of 2-61 inches above the mean for 1904. The 
four largest records in one day were 11-33 inches at 
Evelyn Walley, on October 10; 10-70 inches at Durban, on 
June 1; 10-37 inches at Vogel Vlei, on April 9; and 
10:18 inches at Forestbourne, on October 10. Thunder- 
storms were unusually frequent in December, and prac- 
tically absent in July. The highest temperature recorded 
was 119°-5 at Main, on November 19, and the lowest 
