HUMBOLDT'S COSMOS SIDEREAL ASTRONOMY. 211 



the Canary Isles (the first point at which he touches), he 

 enters into a long discussion on the currents and winds of 

 the Atlantic, that great valley of waters dividing the Old 

 from the New World. The sight of the stupendous Peak of 

 Teneriffe leads him to a dissertation on those various condi- 

 tions of figure of the earth, figure of the object, refraction, 

 &c., which determine the visibility of objects at different 

 distances. Six days at Teneriffe, including an ascent of the 

 Peak, furnish materials for half a volume ; in which are 

 blended geology, botany, and zoology vivid descriptions of 

 the island scenery theories of volcanic phenomena ques- 

 tions as to the temperature and chemical composition of the 

 air at different heights the history of the Canaries dis- 

 quisitions on their discovery by the ancients, and on the 

 origin and primitive language of the Guanches, their earliest 

 known population. Many of these topics have been enlarged 

 or corrected by later research ; but, as handled by Humboldt 

 at this period, they well mark his early vigour and aptitude 

 for such enquiries.* 



Regarding him as a writer merely, this exuberance of 

 knowledge, and his mmm diligentia of illustration, almost 

 pass into a fault, if we might apply the term to qualities 

 thus valuable and rare. Digressions may readily be excused 

 where they bring fresh life and vigour to the subject, and 

 suggest new relations to the mind. But, even under this 

 view, we must consider the tendency in Humboldt's case to 

 be one of excess ; and we notice it the rather from finding 

 proofs of the same discursive method in the work before us ; 

 in which the topics, from their vastness and variety, require 

 constant compression, and an adherence to that proportion of 

 parts which is essential to the unity of the whole. Where the 



* This article having been in great part written during a visit I made to 

 the Canary Isles in 1853, with the volumes of Humboldt in my hands, I was 

 naturally led to the example given above. 



