250
22
Table VIII. shows the mean velocity of the wind. This is greatest in spring when ten tra le- wind and the easterly monsoon co-operate and least in August when the southerly monsoon is blowing. Inversely the daily variation is greatest in August and least in February. The wind is strongest about 1 p., and least shortly after sunset. In winter, at times when the North-east trade-wind blows strongly, its force is often greatest at night, or during the early morning hours. The diurnal variation for January and February is the mean of 9 years only.
Table IX. gives the mean direction of the wind. This veers from ENE in November to SSE in August, but the changes from month to month are slight so long as the North-East monsoon prevails, becoming more rapid as summer approaches. While the change from NE monsoon to S monsoon is moderately gradual, the change from S to NE is very sudden. The wind veers daily from the hour of minimum till the hour of maximum temperature, the daily variation being greatest in summer when the wind is slight. The diurnal variation for January and February is the mean of nine years' observations only.
Table X. gives the mean components of the wind and the resultant force. For January and February the results depend on nine years' observations only.
Table XI. exhibits the four yearly means (1885-1888 inclusive) of total annual amount, duration and velocity of winds from different quarters. Three quarters of it come from E and NE.
Table XII. gives the number of days for each month of the year that the wind came from eight different points of the compass. The great preponderance of easterly wind exists throughout the year though it is less marked during the summer months. The results depend on four years' observations only, 1890-1893.
Table XIII. shows that fogs are common in March and occur also during typhoons in August and September. Electric phenomena prevail throughout the summer months, particularly in July and August. Unusual visibility of distant objects occurs most frequently when the air is cleared from dust by heavy rain in June, July and August. Dew is common in August when also halos and coronas are most frequent. Rainbows are comparatively rare in hot countries where the rain is so heavy and the sun usually too high in the sky. The annual average number of lunar coronce noted during the first five years is 14, while during the last five it is 67. This is due to having the sky more closely watched during the night of late years.
Table XIV. shows cumulus to be the common cloud in China. typhoons and cumulo-stratus during the hottest part of the year. met with during the fine clear weather which characterizes the Monsoon.
Cirrus is most frequent during Small cumulus is most frequently earlier part of the North-east
Table XV. shows the amount of clouds. March is the most cloudy month and the cloudiness steadily diminishes till October (inversely with the amount of sunshine, see Table V) and then increases. On an average there are more clouds at sunrise than about midnight. But looking at the different seasons, it is seen that in summer the amount is greatest in the afternoon and least about midnight, while in winter the amount is greatest during the early morning hours and least in the afternoon.
Table XVI. exhibits the monthly extremes registered. The extreme temperatures are derived from nine years only (1885-1893). The wind velocity in January and February is derived from the same nine years and so also is the solar radiation. The range of vapour tension will be seen to be enormous, The daily maximum of rain occurred on May 29, 1889 when 27.440 inches of rain fell between 6.30 a of that day and the same hour on the 30th. The minimum temperature occurred during the cold wave in January 1893 when owing to the dryness of the air at the same time, water froze readily at the observatory and down to very nearly the sea level and much tropical vegetation was destroyed.
Table XVII. exhibits various data. The barometric tide is large in winter (when the air is dry) and small in summer (when it is damp). The mean diurnal variability of temperature-the mean of the differences of temperature of each day and the next-is greatest in winter. This has been determined from the means of nine years only (1885-1893). The number of days on which at least 0.01 inch of rain fell at the Observatory was a minimum in November and a maximum in June and July. The hourly intensity of rain is greatest in July and least in February. The directions whence clouds in different levels come, together with the wind directions given in Table IX prove the direction of the wind to veer on ascending in the atmosphere. The height of the lower clouds is least in April and greatest in November. The height which it is necessary to ascend in the atmosphere in order to reduce the temperature of the air by one degree is greatest in March when clouds are most prevalent and least in June. It is likewise small in November when they are a minimum. These results depend on eight years' observations only (1884-1891).
Table XVIII. gives the average results of the solar radiation thermometer, the maximum and minimum thermometers, and the mean daily range of temperature founded on the observations of nine years (1885-1893) and also the weight of aqueous vapour in a cubic foot of air in troy grains which depends upon the observations during the entire period.
Table XIX. exhibits 10 yearly means of five-day means.