644 Mr W. HOPKINS, ON THE EXTERNAL TEMPERATURE OF THE EARTH, 



The corresponding heights for one degree centigrade are 



Aug. 17 ... 525,6 feet. 

 Aug. 26 ... 523,2 ... 

 Oct. 31 ... 524,7 ... 

 Nov. 10 ...561,6 ... 

 The first three of these results, it will be observed, are nearly identical, while the fourth is 

 materially different. In these ascents the heights attained above the sea-level were, re- 

 spectively, in feet 



19510, 19100, 12640, 22930. 



The height attained in the third being so much less than those in the other three ascents, it 

 cannot be so well compared as the first two may be with the last. Restricting the comparisons, 

 therefore, to those of the first two with the last, we may easily recognise a principal cause of 

 the discrepancy in the results of the observations. The temperatures of the air, reduced to 

 the sea-level, were, approximately, 



Aug. 17...71°,6, (F.) 



Aug. 26...67°,2, 



Nov. 10 ... 50°,2 ; 



the difference between the last and the mean of the first two being 19°,2 (F.), it might be 

 expected that the decrement of temperature would be less in the last than in the first two 

 ascents. Supposing this to be the true cause of the discrepancy in question, it would appear 

 that a depression of temperature at the surface of the Earth of 20° (F.) would cause an increase 

 of about 20 feet in the ascent corresponding to a decrement of 1°. The mean temperature at 

 the pole is about 80° (F.) lower than that of our own region ; and, therefore, according to this 

 calculation, the increment of height corresponding to a decrement of 1° (F.) at the pole, would 

 be about 80 feet more than in these latitudes. 



It must not, however, be forgotten that there is one unavoidable source of error in the 

 comparison of the results of observation and those of theory on this subject. In the funda- 

 mental equation 



p = a?p (1 + a£), 



t denotes the proper temperature of the air itself, whereas the temperatures observed at 

 different heights above the Earth's surface are those which result from the temperature of the 

 air combined with the all-pervading influence of stellar radiation. The effect of direct solar 

 radiation is, of course, supposed to be carefully obviated. 



Observations on temperature, however, such as those above recorded, enable us, indepen- 

 dently of the error just mentioned, to estimate the effect of an assigned increase or decrease in 

 the Earth's atmosphere ; for if that atmosphere were diminished by a quantity equal to that 

 which occupies a spherical shell immediately surrounding the Earth of 1000 miles, for example, 

 in thickness, the mean superficial temperature of the Earth would be diminished by rather 

 more than 3° (F.), and to a proportionally greater amount if a greater quantity of atmosphere 

 were abstracted. A like increase of temperature would result from a corresponding increase in 

 the quantity of atmosphere. 



