457 



one will attain more or less to the following result: The line 

 •will pass through the southern regions of British North- 

 America, further to the east it will pass through Iceland and 

 Western Norway, thereupon again with a southern direction 

 through Sweden, Finland, Russia and Siberia. Further the height 

 above sea-level will cause variations in the line in the different 

 places. 



Naturallv I have not been able to draw this line even in 



V 



Fig. 5. Pinguicula vulgaris, arctic type. 



Winter-buds and fruit-stalks. Autumn-rosette is not developed. Probably from the 

 end of August. Ilua, Greenland. About -/a natural size. (H. E. P. phot.) 



a tolerably correct manner on the basis of the material at my 

 disposal. For this the difficulties have been too great. In 

 the first place the material has not always been collected in 

 the season most suitable for my purpose; secondly, I have 

 often been obliged to make my investigations on herbarium- 

 plants in which the shoot-structure sometimes is very difficult 

 to determine; last but not least, I have had from the greater 

 part of the regions concerned only a few or no specimens at 

 all for observation. If in spite of these great defects I have 

 nevertheless decided to venture an experiment, ! have done it 

 especially because I think I may be able to give some sugges- 

 xxxvi. 30 



