From my own tables, kept by Newman's best registering thermometers, I find that at Hong-kong, in the months of July and August--the two hottest months in the year--the mercury frequently stood as high as 90°, and one day at 94° Fahr. in the shade. The minimum was generally about ten degrees lower than the maximum. In the winter, from December to March, the thermometer frequently sinks nearly down to the freezing point, and sometimes, although rarely, snow has fallen at Canton and on the adjacent hills. The influence of the sea, however, in this part of the empire, has a tendency to check the extremes of both heat and cold; but these are much greater in the northern interior. The northerly winds in the winter and spring months are severely cold in the south of China; indeed, I have suffered more from cold at Hong-kong and Macao in the month of February, than I have ever done in England.
At Shanghae, in the province of Keangsoo, in latitude 31° 20' north, the extremes of heat and cold are much greater than what we experienced in the southern provinces. Through the kindness of Dr. Lockhart, who kept up my meteorological tables during my absence in different parts of the country, I have obtained a very complete set of observations for nearly two years. From these it appears that in July and August the heat is the greatest; the thermometer in the shade sometimes standing for several days at 100° of Fahrenheit. The heat during these days was almost insupportable to Europeans, who, when I was in Shanghae, were obliged to live in Chinese houses, which, from their construction, were ill calculated to exclude the heat. In the end of October the thermometer sometimes sinks as low as the freezing point. In the evening of the 28th of that month, in 1844, the remains of the cotton and other tropical plants which are cultivated in the fields during the summer, were destroyed by frost. December, January, and February are the coldest months in the year, the cold then being quite as severe as it is with us in the south of England. In the winter of 1844-45 the thermometer sunk as low as 26° Fahrenheit. On the night of the 18th of December, and again on January, the 4th, the index was left at 24°. But that winter, according to Chinese accounts, was peculiarly mild, so much so, that the usual supply of ice could not be procured. In ordinary years the ponds and canals are frequently frozen several inches in thickness, and afford a plentiful supply of ice. I have, therefore, little doubt that in most years the thermometer may be found at least twenty degrees below the freezing point, or at 12° of Fahrenheit, and perhaps even lower. Snow frequently falls, but the sun is too powerful to allow it to lie long on the ground.