506 PROCEEDINGS OF THE AMERICAN ACADEMY. 



Ehode Island, Cumberland (S. T. Olney) : Connecticut, Soutliington, 

 June 4, 1899 (C. ^ Bissell) -, Fairfield, June 23, 1901 (K H. Eames, 

 no. 1G8) : Michigan, Detroit, May 22, 18G4, June, 1860, May 9, 1858 

 (Wm. Boott) : Virginia, Harper's Ferry, May 7, 1881 (John Donnell 

 Smith): North Carolina, Broad River, May, 1841 (i?^<^6? according 

 to Kunze, 1. c). The long slender beak of the perigynium and its 

 essentially northern and montane range suggest that further knowledge 

 of the plant may show it to be well distinguished from C. -pennsylvanica. 

 No other character has yet been found by which it can be recognized, and 

 occasional individuals show transitions in the elongation of the beak. 



Var. VESPERTINA, Bailey, Mem. Torr. CI. i. 74. Rather coarser than 

 the species : the usually very dark staminate spikelet peduncled : peri- 

 gynia more coarsely hairy, almost hirsute. — The northwestern form, 

 from the Cascade Mts. of British Columbia to Oregon and Van- 

 couver Island. 



Carex umbellata. 



Like C pilulifera and C. pennsylvanica, C. iimhellata, Schkuhr, pre- 

 sents considerable variation in the length and breadth of its leaves and 

 in the length of its culms and peduncles. As in those species, likewise, 

 these purely vegetative characteristics in C. umbellata seem to accompany 

 no fixed characteristic of the perigynia, nor any special geographic areas ; 

 and too often the long-peduncled spikelets of the so-called var. vicina 

 may be found on portions of a clump which is otherwise good C. um- 

 bellata. As in the related species just discussed, however, C. umbellata 

 presents at least two geographic tendencies seemingly characterized by 

 constant differences in the perigynia. A third form, of which we as 

 yet know too little, has the perigynia glabrous, thus breaking through 

 one of the distinguishing marks of the Montanae. 



Carex umbellata is related on the one hand to 0. nigro-inarginata, and 

 on the other to C. dejlexa. From these two it is usually distinguished 

 without difficulty, but occasional specimens occur which are perplexing. 

 The writer has found that in such cases the best means of distinction 

 between C. umbellata and C. nigro-marginata is offered by the thickness 

 of the perigynia. In C nigro-marginata the mature perigynia vary 

 from 1.3 to 1.6 mm. in thickness, while in mature C. umbellata they 

 are from 1.7 to 2.4 mm. thick. From doubtful forms of G. dejlexa, C. 

 umbellata may best be distinguished by an examination of the scales. 

 In G. umbellata the scales are nearly or quite as long as the subtended 

 perigynia, while in G. dejlexa they are distinctly shorter. 



