1898-1902. No. 2.] VASCULAR PLANTS OF ELLESMERELAND. 69 



Distribution: East and West Greenland, Arctic American Archi- 

 pelago, Arctic America, Labrador, Canada, Rocky Mountains to Arizona, 

 Alaska, Unalaschka, Northern Siberia, Kamshatka, Baical Mountains, 

 Ural, Novaja Semlja, Spitsbergen, Franz Joseph Land, Northern Russia, 

 Finland, Scandinavia, Riesengebirge, mountains of Great Britain, Faeroes, 

 Iceland, Jan Mayen. 



Saxifraga stellaris, L. 

 var. comosa, Retz. 



S. stellaris jS comosa, Retzius, FL Scand. Prodr., 1779 ; Poiret, in Lamarck, Eneyolo- 

 pedie mSthodique, VJ, ex Ledebour, F1. Ross.; Engleh, Mon. Saxifr.; Langb, 

 Consp. FL Groenl.; Kruuse, List E. Greenl.; Nathorst, N. W. Gronl.; Sim- 

 mons, Prel. Rep. et Hot. Arb.; Kjellman, in Vegaexp.; Macoun, PI. Pribilof; 

 Andeesson & Hesselman, Spetsb. karlv. ; S. stellaris var. prolifera, Sternberg, 

 Mon. Saxifr., suppl. 11; S. foliolosa, R. Brown, Chlor. Melv.; Hooker, F1. Bor. 

 Amer. ; S. comosa, Bhitton & Brown, III FL; S. nivalis forma, Oliver, List 

 fl. pL; S. virginiensis, Hart, Bot. Br. PoL Exp., non Michaux. 

 Fig. Linnaeus, Fl. Lapp., T. 2, fig. 3; Fl. Dan., T. 2354. 



In Ellesmereland, I have alvi^ays seen S. stellaris represented only 

 by this variety, never with even a single flower developed. The same 

 was the case in Foulke Fjord, and Ivsugigsok is the only place north 

 of Melville Bay from which 1 have seen it with terminal flower (in some 

 of the specimens of Nathorst). At Godhavn, it generally had top 

 flowers, even if the others were replaced by gemmae, but in southern 

 Greenland it passes into the typical form, which probably fruits there. 



The var. comosa is doubtless only a form adapted to high arctic 

 conditions, and would, if it were cultivated in more southern regions, 

 probably go over to the typical one. To me, it seems therefore quite 

 out of the question to look upon it as a separate species. 



The variety is first mentioned by Linnaeus, Fl. Lapp. (I have only 

 had access to Ed. II of J. E. Smith, 1792, but this is, in most parts, a 

 reprint of Ed. I). Here a clear description is given, as also a rather 

 good figure, but no name. Retzius, 1. c, p. 79, is the first to use the 

 name ^ comosa, with reference to the figure in the Fl. Lapp., and conse- 

 quently he is to be quoted as author, instead of Poiret who later has 

 used the same name, but who is nevertheless quoted by Engler, 1. c, 



p. 133, and others. 



S. stellaris var. comosa is not common in Ellesmereland and m 

 most localities only single individuals were found, but it is probably 

 spread over the whole country. Hart and Greely indeed have not 

 mentioned it from the northern parts, but the former nevertheless did 

 find it. Specimens in the Nat. Hist. Mus. herbarium show that the "S. 



