THE LAST TEN YEARS. 357 



of the work since the literary aspect is entirely subordinated 

 to the scientific requirements, and the minute exhibition of 

 detail occupies the place of the former generalisations. Of this 

 change Humboldt was himself conscious, and anticipated the 

 verdict of the public. ' As formerly my work was thought to 

 be too meagre and too poetic,' he wrote to Jacobi in 1850, 'so 

 now it will be said to be too redundant and too prosaic. The 

 bold attempt to give well-authenticated facts in astronomy, 

 magnetism, geology, meteorology, and physical geography, in 

 which every detail is worked out as if it were the only point 

 to be considered and discussed until its significance is made 

 clear, deserves, at least, when one is past eighty, some amount 

 of appreciative recognition.' As in discussing the first two 

 volumes our attention was mainly directed towards the literary 

 merit of the work, the task now before us is rather to give a 

 review of ' Cosmos ' in its scientific aspect. There are two 

 passages, however the introductions to the third and fifth 

 volumes to which we shall first direct our attention, as from 

 their import they connect themselves with the 'generalisations' 

 of the former work. 



While still in the expectation that the specific portions of 

 the work would not occupy more than one volume, Humboldt 

 wrote to Bockh on August 10, 1849 : 'My object in the last 

 volume is once more to enchain the reader by the contempla- 

 tion of all that has been attempted during the last 2,500 years 

 towards the solution of the enigma of Nature, and, while 

 touching only 'slightly upon the intermediate steps, to bring 

 out the resemblances and contrasts in the philosophy of 

 science, as developed by Giordano Bruno of the Italian school, 

 by Descartes, and by Newton. The style of the work must 

 again be that of a general survey of philosophic thought, in 

 which the principle of generalisation must not be carried so 

 far as to prevent an interest being aroused in the special and 

 individual points in the systems of the various founders. I 

 admit it is a mannerism, but the renunciation of pathos in 

 my style conceals the intention, and hitherto the mannerism 

 has been successful. The "introduction," from its style and 

 treatment, ought to form a connecting link with the previous 

 volumes.' Nearly ten years afterwards Humboldt forwarded to 



