514 X. GENERAL REMARKS. 



for larger views than those which occupy the minds of 

 the men of species. Unfortunately, men of this latter 

 taste and attainment are quite as necessary to the phi- 

 losophers of science, as the hrick-makers and stone- 

 cutters are necessary to builders and architects. A 

 Humboldt or a Lyell would have made poor progi'ess 

 without the aid of species-describers ; although they do 

 incur the risk of being widely misled by trusting to them 

 too implicitly. 



Filices and Orcliidacece afford opj)Osite examples, in 

 respect to explanation ; while they also contrast oppo- 

 sitely against each other. Their numerical assignment 

 to types here originates almost exclusively in the facts 

 of nature, not in the fancies of botanists. The ferns 

 are either generally diffused through Britain, or are 

 characteristic of the northern and western provinces ; 

 comparatively few being assigned to the English type, 

 and none to the Germanic type. On the contrary, Or- 

 chidacecB are more largely assigned to the English and 

 (especially) the Germanic types ; with a smaller number 

 of widely-diffused species assignable to the British type ; 

 none to the Highland or Atlantic. These assignments 

 are traceable to climate, and perhaps partly to the cha- 

 racter of the ground; "orchids apparently having some 

 special adaptation to cretaceous soils, and many ferns 

 being adapted to rocky sites. 



It is needless to multiply such explanatory examples. 

 Those botanists who have understood the facts and 

 figures of former lists and tables, will be prepared to 

 seek out the explanations for themselves. But it is quite 

 needful to keep in view the very different nature of the 

 table on pages 511, 512, and the three summaries of 

 orders before printed on pages 359 to 367. In the table 

 here under consideration no species is counted twice ; 



