CONTRIBUTIONS FROM THE GRAY HERBARIUM OF HARVARD 
UNIVERSITY — NEW SERIES, No. XLVIII 
I. THE TRUE MERTENSIAS OF WESTERN 
NORTH AMERICA 
By J. Francis Macsripe. 
Tue genus Mertensia reaches its greatest development in the 
United States west of the Great Plains or the Missouri River. No 
broad treatment of the genus has appeared since Dr. Gray’s in the 
Synoptical Flora thirty years ago, although several floras have 
covered the species known to occur within their respective terri- 
tories. These works and the number of Mertensia species recog- 
nized by each are as follows. Piper’s Flora of Washington (Contrib. 
U.S. Nat. Herb. xi. 1906), 14 species; Rydberg’s Flora of Colorado 
(Bulletin 100, Colo. Exp. Sta. 1906), 25 species; Coulter & Nelson’s 
Man. Rocky Mt. Bot. 1909, 14 species and 10 varieties; Wooton 
& Standley’s Flora of New Mexico (Contrib. U. 8. Nat. Herb. xix. 
1915), 8 species. The Synoptical Flora contains 7 species and 3 
varieties, 2 species being confined to the Atlantic states. Since its 
publication (1886), 69 species and several varieties (without includ- 
ing those published here) have been proposed. 
These data have been given in order to show that students are 
agreed in admitting in recent years a large increase to the number 
of species, but that they are not agreed as to what percentage of 
these species are valid. These great differences of opinion are due 
to the lack of any consensus as to what characters have specific 
value. The presence, absence, location or quality of pubescence, 
the relative lengths of corolla-limb, -tube, and calyx, the shape of 
the calyx-lobes and leaves have all been regarded as of equal worth. 
Then habit and general aspect have been considered. I have 
reached the conclusion that, generally speaking, the most salient 
and most reliable of these characters are the shape of the calyx- 
lobes, the location of the pubescence, and, especially when on the 
pedicels, its quality, coupled with aspect and habital characteristics. 
The latter are of great importance, especially in the division of 
the genus into sections, although of course they are characteristics 
1 
