326 Rydberg : Rocky Mountain flora 



Anaphalis angustifolia sp. nov. 



Perennial with a creeping rootstock; stems slender, strict, 

 simple, white-tomentose, 3-6 dm. high; leaves narrowly linear, 

 i-nerved, 5-10 cm. long, 2-5 mm. wide, ascending, densely white- 

 tomentose beneath, less so above, greener and often glabrate in age; 

 inflorescence small, corymbiform, 2-3 cm. high, 3-4 cm. wide; 

 heads hemispheric; involucres about 5 mm. high; bracts pearly 

 white, elliptic, obtuse or acutish. 



This species is more related to the eastern Anaphalis margari- 

 tacea than to A. siihalpina, having narrow and i -ribbed leaves, but 

 differs in the smaller inflorescence, smaller heads, narrower bracts, 

 and narrower and ascending instead of spreading leaves. 



Montana: Mount MacDonald, July, 1900, Elrod & assistants 

 (type, in herb. N. Y. Bot. Gard.) ; Big Fork, Aug. 9, 1901, Umhach 



15- 



Balsamorrhiza 



Professor Nelson has reduced Balsamorrhiza hirsuta Nutt. and 

 B. terebinthacea Nutt. to varieties of B. Hookeri and B. macro- 

 phylla respectively. Although the first has a superficial resem- 

 blance in the cutting of the leaves to B. Hookeri, it is much more 

 closely related to B. macrophylla in every respect, differing only 

 in the smaller heads and more dissected leaves. In his key Pro- 

 fessor Nelson gives the following characters: 



Le&ves entire or somewhat toothed. i. B. sagillala. 

 Leaves not entire, laciniately dentate to bipinnatifid. 



Green, glabrous or sparingly hisute. 2. B. macrophylla. 

 Canescent or lanate. 



The sericeous pubescence appressed or spreading. 3. B. Hookeri. 



The white tomentum often floccose. 4. B. incana. 



How would it be possible to locate B. Hookeri hirsuta by means 

 of this key? As its leaves are pinnatifid, green, and hirsute, it 

 would key into B. macrophylla instead of B. Hookeri. On the 

 following page Professor Nelson gives only the following: "The 

 pubescence roughish hirsute and spreading, not canescent or to- 

 mentose," as distinguishing the var. hirsuta from B. Hookeri, just 

 the same characters which he in the key has used as separating 

 B. macrophylla from B. Hookeri. The latter is not found within 

 the range. 



Balsamorrhiza terebinthacea Nutt. is not closely related to G. 



