Rydberg : Studies on the Rocky Mountain flora 467 



a subspecies of P. caespitosa and makes P. Covillei E. Nels. and P. 

 condensata (Gray) E. Nels. varieties of this subspecies. It is 

 evident that Dr. Brand does not know what P. muscoides is; for 

 the only Montana specimen he cites is Rydberg & Bessey 4815, 

 which belongs to Phlox caespitosa and is so referred in my Flora of 

 Montana. He also cites two specimens from California, Coville 

 2072 from Mount Whitney and Hildebrand from Silver Mountains 

 and also one from Charleston Mountains, Nevada, Purpas 6111. 

 Phlox muscoides Nutt. is unknown to both California and Nevada. 

 Neither P. Covillei nor P. condensata is closely related to P. 

 muscoides. The relationship of P. muscoides is with P. bryoides on 

 one hand, and P. Hoodii on the other; and the calyx is arachnoid- 

 villous, not glandular as in P. caespitosa. 



Dr. Brand made Phlox albomarginata M. E. Jones, P. costata 

 Rydb., P. collina Rydb. and P. diapensioides Rydb. varieties of P. 

 Kelseyi, while he kept P. alyssifolia Greene as distinct and described 

 a new species, P. variabilis, from material which I had included in 

 P. collina. It is evident that Dr. Brand did not know the plants 

 he was so treating. Under his var. albomarginata he gives the 

 following distribution : 



"Montana (nach Jones) — Wyoming: Cooper Creek (Nelson n. 

 4336)." 



Evidently he had not seen Jones's specimens and Nelson 4336 

 is typical P. Kelseyi, and has nothing of the habit of P. albomar- 

 ginata. 



Under each of his var. collina and var. diapensioides he gives 

 the following: "Montana (nach Rydberg)." In other words, he 

 had seen no specimens. Under his var. costata he gives: "Mon- 

 tana: Madison Co. (Nelson 5148)." This number is not found 

 in the herbarium of the New York Botanical Garden, but judging 

 from the rest I am inclined to think that this determination is just 

 as unreliable. It may be that 5148 is a misprint for 5418, which is 

 labeled Phlox Kelseyi, and should be referred to it, but is not 

 typical. It has nothing to do with P. costata. Furthermore, P. 

 costata is not closely related to P. Kelseyi, but intermediate be- 

 tween P. multiflora A. Nels. and P. glabrata (A. Nels.) Brand, 

 but with a densely pubescent calyx. 



In my opinion, P. alyssoides Greene, P. collina Rydb., and P. 



