90 H. G. SIMMONS. [sec. arct. exp. fram 



Parva, dense caespitosa: folia integra, anguste lanceolata vel fere 

 linearia, nervo distincto, pilis rigidis, crassis, ciliata, praeterea glabra 

 vel apicem versus pilis similibus (fiircatisque) insila. Scapi huniiles, 

 aphylli, pilis furcato-stellatis (simplicibusque) obsiti. Inflorescentia pauci- 

 flora, per anthesin subcapitata, demum phis minus elongala. Flores 

 minuli, sepalis angustis, pelalis brevis angustisque. Petala sepalis bre- 

 viora, aequilonga vel pauUo longiora, spathulata, apice rotundata vel 

 retusa, alba. Siliculae primum lanceolatae, deinde rotundatae, crassae, 

 purpureo-brunneae, nitidae. 



D. subcapitata generally forms dense, low tufts. Each rosule of 

 new leaves is surrounded by numerous old leaves, which are so hard 

 in their texture, as to remain for several years. All, however, are den- 

 sely packed together in the contracted stem. Elongated leafy stems 

 such as are represented in the figure of Fries (Till. Spetsb. Fan. Fl., 

 T. 3) are not typical, even though they may sometimes be found where 

 the plant grows among deep moss, or in other dense vegetation. In 

 such individuals also, the leaves will become broader as in the typical 



Fig. a. Tj'pical hairforms from the scape of Draha subcapitata. 



form. I have no such among my material, but I have seen some 

 specimens from Spitsbergen. The "foliis lato-lanceolatis" in Hooker's 

 description of D. micropetala, the only part of it which does not agree 

 with D. subcapitata has reference to a D. alpina form, the original 

 D. micropetala, as already mentioned. The leaves are coarsely ciliated, 

 and similar unbranched hairs are also developed on the uppermost part 

 of the surfaces, especially the upper one. Sometimes some of these 

 hairs become more or less forked, but stellate hairs are not found. The 

 middle vein is very prominent, sometimes quite to the tip of the leaf. 

 The scapes are always leafless, slender at first, later on more 

 coarse and stiff, covered with hairs which may be said to form a link 

 between the forked ones and the starry hairs of D. hirta and D. nivalis. 

 Fries in his description of D. Martinsiana says about the stalk, "pilis 

 minutis simplicibus furcatisque puberulis", but simple hairs are rather 

 seldom found and true forked hairs hardly at all. The same kind of 



