CONTRASTED ISLANDS 83 



straight stick two feet long. Similarly, it is observed in Kam- 

 chatka that those districts that have a sufficient breadth of land 

 from north to south are most prolific in forests and other needful 

 things. On the contrary, the narrower the land gradually be- 

 comes, the more obvious is the change, as for example between 

 the Bolshaya River and [Cape] Lopatka. On the other hand, the 

 region opposite Karaga,^^* where the land is still narrow^er and 

 the latitude, in addition, six degrees farther north, is conditioned 

 quite differently. With regard to the islands situated in the 

 Channel ^8^ the explanation is even more obvious.^^" The fact 

 that the eastern islands, though similarly situated and narrow, 

 were found to be forested has its demonstrable explanation in 

 the circumstance that they lie close under the mainland and 

 thereby are protected, ^^^ while a hypothetical explanation is my 

 conjecture that the corner of America projecting towards Asia 

 decreases in width towards the west and is of the same nature on 

 its northwestern coast as Kamchatka on its northeastern. ^^^ 



188 The narrow neck of the Kamchatka Peninsula in 59° N. (see \'oh 

 I, PI. I). 



189 See footnote 184. 



190 The MS here has the additional clause: "as on their narrow end 

 they are exposed without protection to all the violent winds originating 

 between northeast and southwest." 



191 It is interesting to compare this explanation of Steller's with the 

 similar reflection of modern naturalists visiting the same region. I quote 

 the following from Osgood {North Amer. Fauna No. 24, 1904, pp. 23-24). 

 "Just what are all the causes determining the non-existence of coniferous 

 trees on the greater part of the [Alaska] peninsula can hardly be ascer- 

 tained until more work is done. Possibly one of the most effective checks 

 to the extension of timber southward is the prevalence of wind and storm 

 regardless of temperature. The topography and situation of the peninsula 

 are most favorable for stormy weather. Being long and narrow, with a 

 ridge of high mountains extending throughout its length, and situated 

 as it is between Bering Sea and the North Pacific Ocean, it must neces- 

 sarily receive at nearly all seasons the force of many atmospheric dis- 

 turbances. In the fall it is swept by fierce wunds, whether the temperature 

 be moderate or not. Such conditions would restrict arborescent vegeta- 

 tion in almost any latitude." (S) 



192 The MS here has the following sentence concluding the argument: 

 "For these reasons, then, the islands situated to the east are wooded on 



