Hot and Thermal Springs, 6fC. 263 



If we look at the greatest differences in the third column of 

 figures between places situated at about the same height above 

 the sea, we find that they exceed only three times the above- 

 mentioned difference of 71 feet. This justifies the conjec- 

 ture that these differences are only due to the local influence 

 of climate. At least on the sea coast no other cause can be 

 imagined. The numbers of the third column shew neither an 

 increase nor decrease up to the Metairie dAntisana. It is true 

 the highest number, 775 feet is found near the end, but the 

 next, which is 757 feet, is found quite at the beginning of the 

 series. The case is nearly the same with respect to the mini- 

 mum, 518 feet. The two last numbers but one, 1000 and 

 1151 feet, are, however, astonishingly high, and from this it 

 would appear that the decrease of temperature becomes slower 

 near the limits of perpetual snow. But the last number, 769, 

 is again about equal to the mean. The two last numbers but 

 one, 687 and 758, were found from the differences between the 

 highest and lowest mean temperatures on the level of the sea, 

 and that of the limits of perennial snow. They shew the above- 

 mentioned difference of 71 feet, which may be considered to a 

 certain extent as the maximum effect of climate. 



The means of the four columns of results in the table being 

 so nearly equal, proves that between the tropics, from the sur- 

 face of the sea to the limits of perpetual snow, a constant de- 

 crease of temperature takes place. However, Metairie d'Anti- 

 Sana and Aziifralde Juan must be excepted. 



The mean of all these results gives a difference of height of 

 677 feet for 2°.25 difference of temperature, between the sur- 

 face of the sea and the limits of perennial snow. 



This value only surpasses the mean, which Von Humboldt 

 found from his observations on the decrease of temperature in 

 the air, by 23 feet. It might, indeed, be conjectured that this 

 excess was due to Humboldt's observations having been made 

 in the day, and, therefore, gave a somewhat accelerated de- 

 crease of temperature ; but, in general, no great weight can be 

 attached to this difference, as the numerical elements which in- 

 fluence observations of this sort are very various.* Their cor- 



• The differences of height found from Boussingault's observations, for 

 2°.25 difference of temperature, for example, would have been smaller, if the 

 temperature at the limits of perennial snow should be lower than + 34°.92. 



