1900-1902. No. 2. VASCULAR PLANTS OF ELLESMERE LAND. 73 



Ellesmereland side by the sledging parties. F. flavescens: in clay soil 

 in the great valley at the bottom of the Goose Fjord (3274). 



Distribution: Greenland, Arctic American Archipelago, Arctic 

 America, Labrador, New Foundland, Canada, Rocky Mountains to Colo- 

 rado, Alaska, Arctic Siberia, Arctic Russia, Novaja Semlja, Spitsbergen, 

 Franz Joseph Land, Northern Scandinavia, mountains of Germany (not 

 in the Alps), Jura, Pyrenees, mountains of France, Belgium, Great 

 Britain, Faeroes, Iceland, Jan Mayen. 



Saxifraga groenlandica, L. 

 * exaratoides, n. subsp. 

 S. mixta, Lapeyrouse, F1. Pyren., ex p.? S. exarata, Hooker, FL Bor. Amer., non 



ViLLARS. 



Fig. Tab. nostra 7, fig. 1 — 5. 



Laxe pulvinata, cauhbus numerosis, 1—4 foliatis; folia basalia tri- 

 loba, cuneata, lobis acutiusculis; folia caulina remota, inferiora triloba, 

 superiora linearia, glandulosa; inflorescentia subcorymbosa, flore ter- 

 minali lateralibus majore; flores omnino parvae; lacinia calycina acutius- 

 cula; petala parva, angustata, sepalis parum longiora, alba vel pallide 

 rosea. 



When I first found this plant, I was in great doubt how to classify 

 it. Indeed I did not doubt that it must belong to the form-series of S. 

 groenlandica, but, on the other hand, there was a certain resemblance 

 in the flowers to S. nivalis that was found rather abundantly in the 

 same locality. The small, agglomerate, reddish flowers at first sight 

 reminded me very much of the latter species. I thought also that it 

 could perhaps be a hybrid between the above-mentioned species, but 

 many things tell against that: firstly that its fruit, to judge from the 

 old stalks, was well developed, that it was found in great numbers and 

 that hybridisation is a thing which must not be too easily assumed in 

 arctic regions. As will be shown below, I have since seen specimens 

 of it from other parts of the world also, and have even found it men- 

 tioned in literature, and at last I arrived at the conclusion that it had 

 to be looked upon as a sub-species of S. groenlandica. 



From that species it differs, however, evidently in its very small 

 flowers that are very densely clustered in a sub-corymbose inflorescence, 

 with petals that are not purely white but which have a touch of pink. 

 The flowers also are more numerous than in the common S- groen- 

 landica var. uniflora. The central flower is generally considerably 



