82 The Ottawa Naturalist. [June 



into a short mucro, while in all the other known species, the mid- 

 rib does not protrude beyond the apex of the glume. The species 

 shows in all other respects a close resemblance to A. effusa^ being 

 of low stature with very broad leaves, thick culm, with the bran- 

 ches of the loose-flowered panicle deflexed and, furthermore, by 

 the spikelets being two-flowered and rounded at the base. Hackel 

 (in litteris) placed it nevertheless ifnder Arctophila even if it be 

 somewhat anomalous in this genus on account of the mucro, but 

 as stated by him, it would be still more anomalous in CoLpodium 

 Trin. (as understood by Bentham), because there the midrib never 

 reaches the top of the undivided glume. It would seem as if this 

 interesting addition to the genus Arctophila would warrant its 

 final segregation altogether from Colpodiiuii of Trinius, but strange 

 to say, it is placed together with A, fulva a.n6. pendulina as a true 

 Colpodium by Beal, in his lately published Monograph of the North- 

 American GraminecB. 



And besides this species of Arctophila wi h the flowering 

 glume mucronate, there is, still, another and even more interest- 

 ing type, hitherto undescribed, which we found in the herbarium 

 of the Canadian Geological Survey, which had been collected on 

 Mansfield Island, north of Hudson Bay by Dr. Robert Bell. In 

 this species the flowering glume is distinctly awned, not simply 

 mucronate, a fact that excludes the plant absolutely from Col- 

 podium, while it may be well understood as an Arctophila, and 

 placed next to A. mucronata. It constitutes a species distinct 

 from this not only by the presence of a true awn, but also by the 

 larger number of flowers in the spikelet, the slender culm, much 

 narrower leaves and by an altogether more graceful habit ; we 

 have designated the name A. trichopoda to this species, and a full 

 description and illustration will be published at an early date in a 

 work upon the Hudson Bay Flora. 



The discovery of this well marked species induced the writer 

 to study some more material of the genus as represented in North 

 America, and our investigation has resulted in the separation of 

 three other species, which appear to us as very distinct from those 

 previously described ; they had been identified as A. fiilva and A. 

 pendulina. 



