DERIVATION OF THE FLORA OF HAWAII 19 



positae are carried by the wind, and the fact that the prevailing northeast 

 trade winds would favor their carriage from America to Hawaii, it seems 

 quite as likely that in this way the ancestors of the indigenous Compositae 

 reached Hawaii. 



As to the period of their arrival, assuming that the islands emerged 

 during the early Tertiary, it is hardly probable that at that period the 

 highly specialized and probably relatively modern Compositae of Califor- 

 nia, e. g., Madia, Raillardella, had been developed, and these are assumed 

 to be the nearest relatives of the Hawaiian composites. 



In view of these facts it seems much more likely that the composite 

 element of the Hawaiian flora is relatively modern, instead of being a 

 remnant of an ancient flora. It is now generally believed that the Com- 

 positae are among the most recent plant types, and at present are still in 

 a formative condition. They are not known in a fossil state back of the 

 Miocene, and very few living genera are known as fossils. 18 Their great 

 development and rapid spread in modern times are due undoubtedly to 

 their particularly perfect methods of distribution, either by wind or by 

 animal agency. 



The geological history of California is such as to indicate that most 

 of the members of its extensive and peculiar composite flora are recent 

 developments. 



The extremely specialized Lobeliaceae, which are so characteristic 

 of the Hawaiian flora, are regarded by Guppy as among the early arrivals 

 and, like the Compositae, of American origin. The relationship of these 

 to American forms is, however, by no means so clear as that of the Com- 

 positae. While it is true that the Lobeliaceae are largely developed in 

 America, especially in Mexico and the Andean regions, the family is a 

 cosmopolitan one, and also abundantly represented in the southern hemis- 

 phere. New Zealand has a number of species of Lobelia as well as other 

 genera, and there are numerous species in Australia and South Africa. 



The endemic genera of Lobeliaceae and Compositae comprise about 

 two-fifths of the endemic genera of Hawaiian spermatophytes, and 

 these endemic genera are assumed by Guppy to be the descendants of 

 the earliest immigrants. Of the other endemic genera which he thinks 

 reached Hawaii at a later period, part are believed to be American, the 

 others Malayan. 



The rest of the flora he recognizes as predominantly Australian and 

 Indo-Malayan. 



Guppy is convinced that the most important agents in the distribution 

 of plants throughout Polynesia were fruit-eating birds, and he assigns 



18 Engler and Prantl : "Die Natiirlichen Pflanzenfamilien," IV Theil, 5 Abt. 

 p. 116. 1894. 



