XXVI PRELIMINARY DISCOURSES. 



ttamens, and the pistils. The primary divisions, or Classes, are 

 established mainly on the number, situation, proportion, and connec- 

 tion, of the stamens ; while a large portion of the secondary divisions, 

 or Orders, is based simply on the number of the pistils. Some 

 exceptions occur in the foundation of the Orders : but they are all 

 very intelligible, and readily understood by the student. It is 

 obvious, therefore, that nothing can be easier to acquire, than a 

 knowledge of the Linnean Classes and Orders. The practiced eye 

 even in a flower that it has never before seen determines them 

 at a glance. 



The genus, and species which are the remaining subdivisions are 

 to be ascertained as in all other systems; the former, from the 

 structure of the flower and fruit, the latter, from the peculiar 

 and permanent characters of the residue of the plant. 



It has been often objected to the Linnaean method, that there 

 are too many exceptions to the rules of classification, that a num- 

 ber of genera comprise species which, according to the system, 

 belong to a different Class, or Order, and thus mislead and puzzle 

 the student. This objection is undoubtedly founded on fact ; but 

 the inconvenience has been somewhat exaggerated, and may be 

 greatly remedied, and almost obviated, by simply putting a list of 

 those anomalies at the foot of the respective orders, where the sys- 

 tem would lead us to look for them. Where this is carefully done, 

 it even seems to facilitate the determination of such irregular 

 species, directly pointing them out by name, and designating the 

 genera to which they severally belong. It may be questioned, 

 moreover, whether the exceptions in the definitions of many natural 

 Orders, are not even more numerous than those complained of in 

 the Linnaean method. Instructive, and delightfully interesting, as 

 is the contemplation of the truly natural families, it seems to be 

 necessary to make the definitions of some of them very loose, and 

 comprehensive, in order to embrace all the individuals admitted to 

 belong to them : * so much so, indeed, that unless the student has 



* Take, for instance, the very first family, in Prof. DE CANDOILE'S method of 

 arrangement viz. the order RANUNCCLACEAE ; where we find the following diver- 

 sified not to say incongruous characters, enumerated in the definition of the 

 order: Her bs, or woody vines, with leaves variously dissected, or entire polypetal- 

 ous, or apetalous flowers regular, or irregular sepals 5, 3, or 15 petals 3 to 15, 

 or wanting stamens indefinite, rarely few pistils many, or few, rarely single 

 fruits either dry pods, or akenes, or berries, 1, or several-seeded ! This is, unques- 

 tionably, a correct outline of the features of a natural family sketched, too, by 

 the hand of a master; yet, a Beginner in Botany would probably be apt to think 

 that as a definition, it is rather indefinite, and he would, no doubt, be thankful 

 for any artificial helps that might tend to facilitate his researches. 



