420 On Land-Birds in the Philippines. 



specieSj the varieties of a species, i£ auy, eacli existing in 

 neighbouring but distinct areas. The Great Bronzed Pigeon, 

 Carpophaya miea, has the bronzed shading of the back much 

 deeper in the specimens from Basilan than in those from the 

 central and northern islands, while those from Paragua have 

 tlie wings much bluer in colour. The lied Woodpecker, 

 Chrysocolaptes xanthocejjhulus, from the central islands, has 

 more red on the throat in the specimens from Masbate than 

 in those from Panay and Negros. The Cockatoo and Racquet- 

 tailed Parrot of Mindanao are decidedly smaller than indi- 

 viduals of the same species in the other islands. Other 

 examples of the same kind are numerous^ and there appears 

 to be a tendency in every species to form as many varieties as 

 it inhabits distinct islands with separating sea-channels broad 

 enough to make the passage over difficult and infrequent. 



These facts make it evident that Philippine species and 

 varieties are geographical or local groups depending upon 

 local causes for their existence. They show isolation to be 

 the first and the necessary step in the formation of species. 



The foregoing facts render the belief in the fusion of closely 

 allied species, when thrown together, almost necessary. The 

 volcanic character of the islands and the shallow seas sepa- 

 rating them, with the observed marks of frequent changes 

 of sea-level, compel us to believe that the land-areas of 

 the Philippines have been continually varying, and that, in 

 miiltitudes of cases, closely allied species have been thrown 

 together by the connection of islands formerly distinct. 

 These closely allied species now no longer existing together, 

 they must have disappeared either by the destruction of one 

 or by their fusion. To one who has observed the likeness in 

 size and colouring and notes and food of these allied forms, 

 the latter is the only reasonable hypothesis for the greater 

 number of cases. 



A satisfactory explanation of many or most of the pheno- 

 mena of distribution of genera and species in larger and 

 in continental areas may be found, perhaps, by giving the 

 same prominence to isolation and fusion in the formation 

 of the species occupying them. 



