Vul XIX. I i.ooMis, ['(iriiilimi in /lit .l/ha/rosses ami Petrels. J-^^- 



from im'iigix' scrks tluU similar variations in an\' Albatross arc 

 constant characters. 



Dichromatism is a prominent \'ariation among tlie Petrels, and 

 the basis of a number of apocryplial species in this group ; 

 " Fulmar Its rodgcrsi," an extreme Ught phase of the Fuhnar (/'///- 

 ■iiKiKiis gldcicilis), being an example* (see Plate LIX.) In the Fuhnar 

 and in some other Petrels a difference exists in the geographic 

 range of the phases, but the distribution does not correlate with 

 environmental conditions after the manner of geographic varia- 

 tion, strongly suggesting that dichromatism originated in saltations 

 (mutations). The phases of the Giant Fulmar {Macronecles gi- 

 gantens), of the Neglected Petrel {Plerodroma neglccla), and of the 

 Wedge-tailed Shearwater [Piiffinus chlororhynchus)-f arc additional 

 instances of dichrcmiatism in the Petrels. It is particularly 

 significant tliat a light and a dark phase occur in the downy 

 young of the Galapagos Albatross. J To what extent dichromatism 

 prevails among the Albatrosses and Petrels can be determined only 

 l)y an exhaustive study of large series from numerous breeding 

 stations. It is highly probable that the so-called intergrading 

 species, like Pitffinus assUnili's, will prove to be dichromatic 

 variations. 



Geograpliic or environmental variation has long been a stumbling- 

 block in the way of systematists. In America, during the last 

 centur}/, many conspicuous variations were mistaken for constant 

 characters and described as species. Later, when larger series 

 were available for study, these alleged species were discovered to be 

 imstable and were demoted to the rank of varieties. Subsequently 

 the term variety was supplanted by the term sub-species, it 

 being taken for granted that geographic variations are incipient 

 species. Such is the genesis of the sub-species in America. 

 Individual, dichromatic, and geographic variations all have evo- 

 lutionary possibilities, but we do not know the remote future of 

 any of these variations, nor the manner in which existing bird- 

 species were evolved. It follows, therefore, that we merely trade 

 on our ignorance when we affirm that geographic variations are 

 the beginnings of future species. Often has it been asserted 

 in their defence that sub-species are aids in the determination of 

 migration routes. Words alter no facts. It is the fact of 

 geographic variation, and not the trinomial, that aids in the 

 determination of migration routes. The White-faced Pi>trel 

 {Pelagodronia marina) and Kuhl's Shearwater {Piiffniiis kuhli) are 

 examples of Petrels exhibiting geographic variations. 



The term species, from the standpoint of birds, has been used 

 frequently in the foregoing paragraphs. Two meanings are in- 

 volved in tlie definition of this term. One pertains to nature, 

 the other to classification. In natur(\ a bird species is a fund- 



* Cf Proc. Calif. Acad. Sci., 4th series, vol. ii., part ?., 1918, p. 88. 

 t Cf. Auk, vol. xxxvi., 1919, pp. 487-489, plate xix. 

 X Cf. Proc. Calif. Acad. Sci., 4tli series, vol. ii., part 2, 1918, pp. 78, 79, 

 plates ix.-xi. 



