444 ALPINE AND ARCTIC PLANTS 



mountain^ and the same remark applies to all the plants that 

 one after another cease to appear at various heights from the 

 lower valleys. One by one they become stunted and depau- 

 perated, and then cease, without any semblance of an attempt 

 to vary into new and hardier forms. And this must have been- 

 proceeding, be it observed, from all those thousands of years 

 that have elapsed since the elevation of the mountains out of 

 the glacial seas. It is to be observed, also, that the new plants 

 that occur in ascending, often belong to different genera and 

 families from those left behind, not to closely allied species; 

 and in the few cases in which this last kind of change occurs, 

 there is no graduation into intermediate forms. For instance, 

 Solidago thyrsoidea and S. virga-aurea ^ occur around the base 

 of the mountain, and for some distance up its sides. At the 

 height of four to five thousand feet the latter only remains, 

 and this in a dwarfish condition. This corresponds to its dis- 

 tribution elsewhere, for, according to Richardson, it occurs in 

 lat. 55° to 65° in Arctic America, and according to Hooker, it is 

 found in the Rocky Mountains, while it also occurs in the hills 

 of Scotland, and very abundantly in some parts of Norway. 

 In the White Mountains S. thrysoidea prevails toward the base, 

 S. virga-aurea toward the summit ; and at the top of Tucker- 

 man's ravine I found the former of these golden rods in blos- 

 som, within a few hundred feet of the latter, each preserving 

 its distinctive peculiarities. Much has lately been said of the 

 appearance of specific diversity that results from the breaking 

 up of the contini^ity of the geographical areas of plants by 

 geological changes ; but here we probably have the converse of 

 this. The mountain species is no doubt a part of the older 

 Arctic flora, the other perhaps belong to a more modern flora, 

 and they have met on the sides of the White Hills. 



^ Macoun thinks that most of the specimens referred to this species be- 

 long to the allied form, S. Mulltinallata^ Ast, which is very extensively dis- 

 tributed on the mountains of British America and in the Arctic regions. 



