1916] Fernald, — Some Allies of Antennaria alpina 237 



0.1 mm. broad, much broader than the acute intermediate ridges. — 

 Newfoundland: limestone barrens near sea-level, Pointe Riche, 

 August 4, 1910, Ferncdd & Wiegand, no. 4,139 (type of A. alpina, 

 var. cana, in Gray Herb.); dry limestone barrens, tipper slopes and 

 tablelands, altitude 200-300 m., Table Mountain, Port a Port Bay, 

 August 10, 1910, Ferndd & Wiegand, no. 4,1 41, July 10 & 17, 1914, 

 Fernald & St. John in Plantar Exsiccatae Grayanae. 



When A. alpina, var. cana was published no clear characters were 

 recognized by which to distinguish it specifically from A. alpina (L.) 

 Gaertn. and specimens with canescent foliage were cited from Norway, 

 Greenland, Labrador and Newfoundland. The recent collection of 

 material, consisting of several hundred individuals, brought back from 

 Table Mountain by Mr. St. John and the writer is so consistent in its 

 characters and so different in some important points from A. alpina 

 that a more detailed study of the group has been undertaken. A. 

 alpina of northern Europe and Greenland and its var. canescens Lange, 

 Fl. Dan. xvi. (fasc. xlvii), 9, t. 2786 (1809) — overlooked when var. 

 cana was published — have the oblanceolate rosette-leaves terminated 

 by a distinct short mucro, but in all the Newfoundland material the 

 shorter more cuneate-obovate leaves are blunt. In A. alpina and var. 

 canescens the denuded mature receptacle exhibits only 20-30 large pits 

 0.3-0.4 mm. broad, but in the Newfoundland plant, A. cana, the pits 

 are very numerous (00-100) and much smaller, about 0.1 mm. broad. 

 In A. alpina the cauline leaves are larger, the lower often 2 cm. long 

 by 3 mm. wide, and commonly fewer (4-9). It seems wiser, therefore, 

 to treat the plant of the Newfoundland limestones as a distinct species. 



The specimens from Norway and Greenland, and two of those from 

 Labrador originally cited under A. alpina, var. cana belong, appar- 

 ently, to A. alpina, var. canescens Lange. But one of the Labrador 

 plants so far departs from A. alpina, on the one hand, and A. cana, 

 on the other, as to seem specifically separate from both. In this 

 plant, collected by Mr. Sornborger at Rama, the rosette-leaves are 

 like those of A. alpina, var. canescens, but the flowering stems, instead 

 of being flexuous as in A. alpina and in A. cana, are stiffly erect, with 

 crowded leaves; and the receptacle of the Rama plant resembles that 

 of A. cana in its numerous tiny pits, but the intermediate ridges, 

 instead of being acute, are rounded and as broad as the pits. This 

 plant seems heretofore to have been unrecognized and it may be called 



Antennaria Sornborgeri. Planta humifusa ramis prostratis sub- 

 ligneis 1 dm. longis, stolonibus confertis perbrevibus; foliis basilaribus 



