316 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1925 



In spite of strong normal currents at right angles to the favorable 

 direction, seeds are occasionally floated to Oahu by abnormal cur- 

 rents, and the tree is now establishing itself there, niore than 25 

 miles from Molokai." 



The facts that most of the lowland plants and animals of the 

 wide stretch from Fiji and the Carolines to the Low Archipelago 

 and Hawaii are of the same or closely allied species, and that at 

 the same time there is a progressive dropping out of species east- 

 ward, suggest strongly that the distribution of life forms has been 

 from scattered island to island and has been accomplished by trans- 

 porting agencies that are rather efficient, though not perfectly so. 

 The fact that relatively few forms are of American origin (except 

 remotely) suggests the inadequacy of the normal trade winds and 

 normal currents as agencies of dispersal, in accord with Wallace's 

 law concerning normal winds and currents. However, the relatively 

 small part that South America has played in supplying forms to the 

 Pacific Islands doubtless is due partly to two special conditions. 

 One is found in the few islands in the eastern part of the tropical 

 Pacific. Indeed, good atlases show no islands between the Low 

 Archipelago and South America, about 3,000 miles awa3^ The Gala- 

 pagos Islands, on the Equator, are almost the only islands north of 

 that zone in similar longitudes. South of 24° S. latitude are only 

 Easter Island (28° S.), Sala-y-Gomez (261/2° S.), and a few other 

 islets or rocks to the west and northwest of Easter Island, and also 

 San Felix and Juan Fernandez Islands near the eightieth meridian, 

 not very far from South America. 



The second special condition highly unfavorable to the spread of 

 South American forms over the Pacific Islands is the fact that much 

 of the western portion of tropical South America is almost barren 

 of life, on account of the extreme aridity of the lowlands and the 

 presence of the lofty Andes only a short distance from the coast. 

 Somewhat similar conditions obtain over a wide belt in the North 

 Pacific. There are almost no islands between Hawaii and North 

 America, and most of the coast of Mexico is almost barren on account 

 of aridity. On the other hand, the southeastern coast of Asia and 

 the East Indies teem with forms adapted to the climatic and soil con- 

 ditions which obtain in most of the Pacific Islands. 



Another argument of advocates of great extensions of the lands 

 is the following : They say that wind and currents, even those asso- 

 ciated with hurricanes, apparently are not effective in the dispersal 

 of certain types of life, as shown by the absence from the coast of 



" Oral communication by C. S. Judd, Territorial forester. Several illustrations of the 

 significance of changes in ocean currenta in the Atlantic are given in papers In the Pro- 

 ceedings of the First Pau-Paclflc Scientific Conference, vols. 1 and 2. 



