DE CANDOLLE'S PHYTOGRAPHY. 293 



or region, are discussed in chapter X. In Floras, as in more 

 general works, abridged descriptions or diagnoses suffice, 

 indeed are preferable in all cases where the region has 

 been pretty well explored, and where the materials can be 

 thoroughly elaborated. Formerly all considerable Floras were 

 written in Latin, at least the characters. So they would con- 

 tinue to be if the convenience of botanists and the advance- 

 ment of science only were to be considered. But Floras are 

 used by many to whom even Linnsean Latin would be a 

 stumbling-block. Fortunately the difference between good 

 botanical English and botanical Latin is not wide, and will 

 not seriously trouble a French or German botanist. The 

 converse hardly holds. The greatest Flora written in English, 

 we might say the best great Flora in any language which has 

 ever been produced and completed, is Bentham's "Flora 

 Australiensis," in seven octavo volumes. Touching upon 

 works of special illustration, the " Botanical Magazine " is 

 justly singled out for praise, for sustained botanical correctness 

 under difficulties, and for its great influence uj)on the science. 



De Candolle insists much on the importance of describing 

 and well classifying the varieties of a species, and of distin- 

 guishing them as much as possible into grades, such as sub- 

 species or races, varieties, subvarieties, etc. We suggest that 

 this can be done with great advantage only when the forms 

 are comparatively definite, or have been described as species. 

 We think that only the more salient and definite varieties 

 should be distinguished by names ; otherwise the names and 

 the groups will be limitless. 



In the eleventh chapter, on partial descriptions of groups 

 from the point of view of organography (a term which our 

 author prefers to morphology), of physiology, botanical geog- 

 raphy, etc., our author has some pertinent remarks upon the 

 helps which all such studies are offering to phytograph}^ 

 which will gradually extend its domain over them ; and upon 

 the obvious advantage and great need of having results of 

 histological researches expressed descriptively, under some- 

 thing like a common terminology, and with due regard to 

 rules which have governed the more matured branches of 



