1823.] M. Humboldt on Volcanoes. 107 



the melted substances proceed from the interior of the earth to 

 its surface, the more important is it to become thoroughly- 

 acquainted with that structure, by exact measurement. The 

 interest attached to this measurement, which has been a parti- 

 cular object of my examination in another part of the world, is 

 heightened by the consideration, that that which is to be mea- 

 sured is a variable magnitude. The physiognomy of nature con- 

 sists in the change of phaenomena tending to connect the present 

 with the past. In order to ascertain a periodical return, or the 

 laws of progressive natural changes in general, certain fixed 

 points are necessary ; and observations carefully made at stated 

 periods, may serve for numerical comparison. Had the mean 

 temperature of the atmosphere in different latitudes been 

 observed for a few thousand years, and the mean height of the 

 barometer at the level of the sea, we might now know in what 

 proportion the heat of different climates has increased, or dimi- 

 nished, and whether the height of the atmosphere has undergone 

 any changes. Similar points for comparison are required, for 

 the variation and the declination of the magnetic needle, and 

 for the intensity of the electromagnetic power, upon which two 

 excellent philosophers of this Academy have thrown so much 

 light. If it be a praiseworthy undertaking of learned societies 

 to inquire assiduously into the changes of temperature undergone 

 by the globe, into those which take place in the pressure of the atmo- 

 sphere, and in the magnetic variation, — it is the duty of a travelling 

 geognost, in ascertaining the inequality of the earth's surface, to 

 consider, principally, the variable height of the volcanoes. What 

 I formerly attempted on the mountains of Mexico, on the Toluca 

 Nauhiampatepetl and Xorullo, and in the Andes of Quito, on 

 the Pichincha, I have found opportunity, since my return to 

 Europe, to repeat at different periods on Vesuvius. Saussure 

 measured this mountain in 1773, at the time when both sides of 

 the crater, the south-eastern and north-western, appeared to be 

 of equal altitude ; he found their height to be 609 toises (3894 

 feet) above the level of the sea. The eruption of 1794 occa- 

 sioned a fall on the south side, wliich even the unaccustomed 

 eye discovers at a great distance. In 1805, I measured Vesu- 

 vius three times, in conjunction with M. von Buch, and M. Gay- 

 Lussac ; we found the elevation of the northern edge, opposite 

 to Monte Somma, la Rocca del Palo, to be exactly the same as 

 Saussure had before determined it ; the southern edge we found 

 71 toises (454 feet) lower than it was in 1773 ; the total height 

 of the volcano on the side opposite Torre del Greco (towards 

 which side the fire seems to have acted the most powerfully, 

 during the last 30 years), had diminished one-ninth part. 



The cone of ashes on Vesuvius bears the proportion of one- 

 third to the height of the whole mountain, that on Pichincha is 

 as 1 to 10, and that on the Peak of TenerifFe as 1 to 22 ; Vesu- 

 vius has, therefore, the largest cone of ashes in proportion, 



