534 CLASSIFICATION AND DIVISION OF 
The adoption by Willdenow of the genus Elichrysum of 
Tournefort, so far from lessening, only served to increase 
the confusion previously existing. Willdenow assumed, as 
the basis of his genus, an erroneous character, namely, 
receptaculum nudum,'''' which can only refer to a very 
small portion of his Elichrysa. Our distinguished coun- 
tryman Mr Brown was the first to attempt a proper limita- 
tion of these genera, in a paper entitled, " Observations on 
the natural Family of Plants called Composite,'''' inserted 
in the 12th volume of the Transactions of the Linnean 
Society. The reading of this admirable paper first led me 
to an examination of these genera, and the divisions I shall 
have to propose in the sequel, are the result of that investi- 
gation ; but convinced that the subject is not yet exhausted, 
I give some of these divisions, with considerable diffidence, 
and I shall feel happy, if such of my botanical readers as 
possess the opportunities, will extend the research, and con- 
firm the observations which I have made. 
Before entering on the particular subject of this memoir, 
it will be perhaps advantageous to give some remarks illus. 
trative of Compositte in general. The Compositae of all 
known families of plants are the most natural, and the most 
extensively distributed over the surface of the globe. In the 
Flora of almost every country, whether situate without or 
within the Tropics, they are found to constitute nearly a 
tenth of the Phasnogamous Vegetation *. From this exten- 
sive distribution of the Compositce has resulted the apparent 
necessity of dividing them into divers separate groups or 
families ; and hence we find in the writings of the earliest 
systematic botanists, from Csesalpinus down to the present 
time, that various attempts have been made to divide theril. 
* In tropical countries possessing great diversity of surface, such, for ex- 
ample, as South America, the proportion is considerably greater. 
