OBSERVATIONS TAKEN IN INDIA. 
325 
The first feature is the close approximation of the annual means at Madras for the 
four years ; the difference between the greatest and least being only l°-26 Fahr., 
although the difference between the means of the same month in successive years 
may have exceeded three degrees. Nevertheless the means of most of the months in 
successive years show a great amount of uniformity. The coldest months are De- 
cember or January, and the hottest May or June. In 1842 and 1843 December was 
the coldest month. In 1844 and 1845 January was the coldest. In 1842 and 1845 
May was the hottest month. In 1843 the maximum mean heat was not reached until 
July, and in 1844 it occurred in June. Although these discrepancies occur, the 
monthly increment or decrement to or from the maximum period is gradual, and 
rarely checked or inverted. In Bombay the annual mean temperature for 1843 is 
almost identical with that of Madras for 1843, although the two places differ 5° 41' 
in latitude. The monthly means of the two places correspond tolerably well from 
January until July; for instance, February corresponds within a tenth of a degree, 
and June within two-tenths. But in July there is a difference of 3°‘79, xlugust 3°'31, 
September 3°’ 16, Madras being plus: then the sign changes and Bombay becomes 
plus for the remaining months of the year, and also in January. The coldest month 
in Bombay was January, and the hottest May. In Bombay, in 1843 and 1844, unlike 
Madras, there is an inversion in the decrement of heat on the sun’s going south ; for 
October is represented as hotter than September in both years: a very remarkable 
fact is exhibited by these hourly mea'ns, namely, that neither the south-west monsoon 
at Bombay, nor the north-east monsoon at Madras, at all affect the motithly mean 
regular increment or decrement of heat, with the exception of October in Bombay. 
It will be observed that the maximum heat, both at Madras and Bombay, was not 
when the sun was vertical at either place passing to the northern tropic ; but 
when the sun was near to the northern tropic in June, the mean temperature in 
1843 at Madras, in lat. 13° 4', was at its maximum in July, while at Bombay, in 
lat. 18° 55', the greatest heat was in May of that year ; and though the sun passes over 
both places again to the south the mean temperature gradually declines in each year, 
with the exceptions noticed. I was desirous of inserting in the Table of Mean Tem- 
peratures those for 1843 and 1844 at Calcutta, for comparison with the Madras and 
Bombay observations for the same year ; but on referring to the volumes of the 
Bengal Asiatic Researches for 1843 and 1844, I found that the tables had not been 
inserted. They were met with however in the Journal of the Horticultural Society 
of Calcutta, but proved to be records of the temperature during the day only at 
9^ A.M., noon, 4 p.m. and sunset*. A mean temperature from such data would neces- 
* In the fifth volume of the Calcutta Journal of Natural History, Dr. M'Clelland states that the mean 
temperature of 1844, from daily observations, 'was 82°'35, the coldest period at sunrise, and the hottest at 
2** 40™ P.M. daily; the maximum heat 104° on the 10th of April, the minimum 51°’7 on the 19th of January, 
and the annual range 52°'3. He speaks also of the minimum daily pressure of the barometer occurring 
twice at 6 p.m. in January, February and May. 
