PREFACE 
of the variable genus Roberisonia; retaining ne- 
vertheless in both tribes, some of the leading va- 
riations as named varieties, or subspecies, 
It may here be imagined by some, that the au- 
thor's opinions concerning the specific distinctions 
of plants, are as peculiar as those above advanced 
respecting the genera. — For his continued experi- 
ence seems at length to confirm him in the doc- 
trine which he has ever in private proposed ; but. 
which now he ventures thus publicly to anneunce, 
viz. That there are, on the one hand, many native 
plants which afford zo spécifie characters, in the 
Linnean sense of those words, and which still are 
species: and that, on the other hand, there are 
some, which, although they present abundant s 
pe- 
cifie characters of the Linnzan kind, are never- 
theless varieties only. 
Hence it is conceived that the Hyessidzon Saxi- 
frage of. Snowdon, &c. are but seldom, if ever, 
exactly the same. as those of Scotland: or that 
either precisely resemble those which inhabit the 
mountains of Kerry. . And the keen eye of a bo- 
tanist, by their ^habit alone, is able to distinguish 
the one from the other, however trifling their ac- 
tual discrepancies. may be. 
»sfThe-4 an tribes are perhaps more diffi- 
eili to understand correctly than any other ; for 
OR dm the eye of a botanist may be able to dis- 
eriminat ongst them, the tongue often, and 
even. vs pen, is foiled in attempting to describe 
what is actually perceived.. I In this intricate de- 
partment. fecere above all others, the author 
