STUDIES IN THE COMPOSIT.E. 269 
Mr. Wiegand's description of his B. dentata reads as if he 
had ineluded this most remarkable plant under that. But 
it can not in any reasonable probability have formed a part 
of Nuttall’s B. quadriaristata var. dentata; indeed Mr. Wie- 
gand's diagnosis actually excludes Nutall’s plant, the essen- 
tial character of which is, according to Nuttall, its dentate 
leaf-margin. But the leaves of B. dentata, Wiegand, are to 
be either serrate or incised or parted, a descriptive phrase 
which excludes the term dentate. Judging from plants 
extant in herbaria, such as Mr. Howell’s from Sauvie's Island, 
which come from very near Nutall’s station, the true dentata 
variety of that author is allied to B. cernua by its strongly 
nodding heads and quadriaristate achenes, and is by far 
smaller than any of the foregoing new species, and has its 
leaves commonly nearly dentate, though now and then 
exhibiting a margin that is between serrate and dentate. 
Such a plant, inhabiting as this does the seaward slope of 
Oregon, Washington and British Columbia, can rationally 
be accepted as iu all probability identical with Nuttall's 
variety dentata ; and there is no other known that can. 
This practice of adopting an antiquated varietal name for 
a proposed new species without taking any pains to make 
sure of the identity of the old variety and the new species is 
a most reprehensible one, as tending only to instability in- 
stead of stability in names. 
I would remark further, that the assuming Nuttall's 
variety dentata and Torrey & Gray's variety elata to be one 
and the same, as is done by Mr. Wiegand, seems quite 
gratuitous. I should readily predict that, if ever it is ascer- 
tained just what the two plants of old so named are, they 
wil be found to represent each a good species Both 
names had much better been left unused; for by such 
neglect of them confusion would have been less confounded. 
