VOL. 5] New Species of Galium. 57 
surface, glabrous on the lower; flowers axillary and terminal, in 
clusters at the ends of the branches, apparently yellowish, corolla 
four-parted, pedicels short; fruit not seen. 
A single plant, San Francisco, Cal., collected by Mrs. Brande- 
gee, July 4 (no year given). G. chartaceum, is another anomalous 
species to be placed provisionally near G. Miguelense, though like 
G, muricatum it is not closely related to that species. 
A few of the foregoing descriptions, such as are based upon a 
small amount of material, may have to be modified somewhat 
when more specimens are available, but they have been formu- 
lated only after a careful examination of a large number of sheets 
of W. American Galiums. Nearly all of the Western species of 
Galium are so variable that to base descriptions upon the extreme 
of any one might exclude many forms which upon the examina- 
tion of considerable material are shown to properly belong to the 
species. Even species common to both the Eastern and Western 
States, and which are fairly well defined in the East, become ex- 
ceedingly variable in the West. Some forms of G. trifidum Pacifi- 
cum, Wiegand, and G. tinctorium submontanum, approach each 
other very closely. G. trifidum Pacificum, however, almost in- 
variably has a three-parted corolla, while G. tinctorium submon- 
tanum has the corolla either three- or four-parted and also has a 
larger number of leaves in the whorl. The whole trifidum group 
is certainly composed of very closely related forms, and G. ¢incto- 
rium is considered as a species here simply because in the East it 
is fairly well marked and it seems more convenient to so consider 
it. G. trifidum is probably more variable than G. tinctortum, but 
the varieties of both species in the East as well as in the West 
-must be considered as approaching each other very closely in 
some instances. 
G. cymosum, Wiegand, and G. arcuatum, Wiegand, formerly in- 
cluded under G. ¢rifidum, the writer has not seen, but from the 
descriptions given in Torr. Bull., xxiv., they would appear 
to be sufficiently distinct from other forms of the “7jidum group 
to be retained as species. — : 
