Macbride — Revision of Oreocarya 21 
next work of importance in this connection was by Dr. Rydberg 
in his Flora of Colorado, Bull. 100, Colo. Agric. Exp. Sta., 286 
(1906), where he keyed out nineteen species. Miss Alice Eastwood, 
in a paper devoted to this genus, Bull. Torr. Club. xxx. 238-246 
(1903), contributed largely to this increase in the number of recog- 
nized species. Finally the treatment by Professor Aven Nelson in 
the Coulter-Nelson Man. R. Mt. Bot. 416-419 (1909) gave a key 
and also descriptions with citations, recognizing nineteen species 
for the central Rocky Mountain region. Since the genus reaches 
its greatest development in Colorado I am most indebted to these 
works, but floras covering the outskirts of the range of the genus 
have been helpful for their regions, notably Wooton & Standley in 
Contrib. U.S. Nat. Herb. xix. 544-546 (1915) and Piper, Contrib. 
U.S. Nat. Herb. xi. 481-482 (1906). When one considers that 
sixty species have been credited to this genus although only nine- 
teen have appeared in any one work, the need of a general revision 
is apparent. It is of interest that this remarkable increase from 
seven species in 1885 to sixty in 1916 has been largely warranted, 
having been due at least in great part, to the discovery that the 
species possess excellent characters of fruit, which may serve to 
distinguish them when other characters are not apparent. In this 
respect Oreocarya resembles greatly Cryptantha. Other characters 
that are important are the shape of the calyx-divisions, nature of 
the pubescence and the inflorescence. Habit, size of corolla and | 
duration are often to be considered. The color of the corolla is 
constantly white, yellow, or white and yellow. Compared with 
Mertensia the genus furnishes some striking contrasts. For in- 
stance, the characters are usually perfectly definite and the species 
rarely exhibit perplexing variations. This may be illustrated con- 
cretely by the fact that, although forty-five species of Oreocarya are 
recognized only three have been noted as varieties while in Mer- 
tensia, the recognition of thirty-two species has disclosed nineteen 
variations that seemed to be worthy a name. The genus, there- 
fore, is not difficult provided the specimens are in fruit. But it is 
impossible in many cases to determine accurately specimens that 
are only in flower. Collectors, therefore, should use the same effort 
and care to secure mature plants as in the genus Cryptantha. 
A very promising and interesting field for investigation, in this 
group especially, is a study of the degree to which certain species 
