554* Reduction of the Meteorological Register, 



It remains to notice the gradations of the mean temperature from 

 one month to another, as exhibited by the foregoing Table. The 

 results bear on the peculiar effects of the cold season on the constitu- 

 tion. The difference of temperature between October and November 

 was 3.80, and between February and March 8.23, more than double 

 the amount of all the changes put together from April to Septem- 

 ber. The following exhibits these results. 



Between the Mean Temp, of Oct. and Nov. — 3.80 difference. 



— 



November and December, — 6.49 



— 



— 



December and January, . . + .24 



— - 



— 



January and February, . . -f- 6.05 



— , 



— 



February and March, . . + 8.23 



— 



— 



March and April, . • + 3. 



— 



— 



April and May, .. .. — 1.92 



— 



— 



May and June, . . .. — 1.87 



_ 



— 



June and July, .. — 1.38 



— - 



— 



July and August, . . — .65 



— 



— - 



August and September, . . -j- 1-73 



— 



— 



September and October, — 1.80 



— 



The wet bulb thermometer would add greatly to the value of these 

 results, but there are discrepancies in the register of the instrument 

 in use, which prove it to be imperfect. With this exception, the 

 observations are highly creditable, and have now we believe been 

 continued for a sufficient series of years to render their results of the 

 highest importance. 



Without drawing any conclusion from the results of a single year, 

 particularly when we have materials for a series of years available, 

 we may advert to the remarkable correspondence between the ratio 

 of Variation of temperature, and that of Mortality, as exhibited in 

 the following Table. The third column is taken from an important 

 paper by Colonel Sykes, read at the recent meeting of the British 

 Association, at York. 



