182 NOTICES or BOOKS. 



Chapter 15. On the Origin and Dui-atiou of Genera : and the changes 

 that then' habitats are undergoing at the present epoch. 



Chapter 16, On the Geographical Position of Families: on their 

 limits, and on the general character of the countries they occupy. 



Chapter 17. On the Distribution of the Plants belonging to a Family 

 in the area the family occupies, and on the comparison of different 

 Families from this point of view. 



Chapter 18. On tlie Area occupied by Families. 



Much curious matter is introduced into several of the above Chapters, 

 though the subjects to which they refer are not treated in the same 

 manner or spirit that the previous ones are. They are all of too frag- 

 mentary a character to admit of any general conclusions being drawn 

 from them ; they are also so much broken up into articles with dif- 

 ferent headings, under many of which the same kind of information is 

 repeated in reference to the speciality of which the article treats, as to 

 be rather confusing. In those especially devoted to the distribution of 

 Natural Orders it appears to us that the extremely different opinions 

 entert-ained by botanists as to the limits of the Orders themselves is an 

 obstacle to any results being obtained, except from 'a tabulation of the 

 genera and species of the whole vegetable kingdom upon two distinct 

 plans ; one under the fewest Natural Orders under which they can be 

 arranged, and the other under the greatest number into which they 

 have been divided. 



M. de Candolle, however, probably conceives that the usually ac- 

 cepted Natural Orders are capable of better limitation than we do. 

 This is in accordance with his theory that genera are more natural 

 assemblages of species than species are of individuals, and that the 

 higher we ascend in the scale the better limited are the groups. Now, 

 though it is very true that although in the present state of our know- 

 ledge the structural and physiological distinctions found to exist be- 

 tween Phaenogams and Cryptogams are absolute, those between Monoco- 

 tyledons and Dicotyledons very nearly so, and that Gymnosperms and 

 Angiosperms are well structurally marked, we do not find that the Na- 

 tural Orders of these great divisions are so well defined. It is true that 

 they may be perfectly natural for the most part ; that is to say, that the 

 osculant genera may be few; but when few, they often embrace a host 

 of species: and though it may be by but one genus that an Order is 

 united with another Ord(;r, the majority of the genera in that Order 



