Todd : A Revision of the Geni ChjEmepelia. .">l>;$ 



Phytogeny, — Even in i In- case oi insular forms intergradation with 

 tin- races occupying adjacenl areas is mure or less evident, and in the 

 few instances noted beyond, where this is apparently not the case, it 

 i- probable thai further explorations will bring t<> lighl such inter- 

 mediate examples. We are justified, therefore, in attributing to 

 C. passerina a practically continuous and verj extensive distribution, 

 stretching from Peru and Brazil northward through Central America 

 and Mexico to southern Texas and Arizona, and through the Antillean 

 chain to the southeastern United States. Being a bird of the open 

 country, there are of course wide forest areas from which it has not 

 been reported, but otherwise the only bar to its spread would seem 

 to be the higher mountain ranges. We may thus account for the wide 

 dispersion of the Brazilian form, C. p. griseola, as well as for the fact 

 that there are no less than three distinct races in the Andean region 

 of Colombia, corresponding to three separate physiographic areas. 

 The species as a whole is undoubtedly of South American origin, but 

 which of the existing races is nearest the primitive form is a question. 



Were the various races dependent upon a single set of characters for 

 their definition, the tracing of their genetic relationship would possibly 

 be simple enough, but the matter is vastly complicated when varying 

 combinations of several different characters, not correlated with each 

 other, are involved. To illustrate: take the case of three forms which 

 agree in having the base of the bill yellow or orange-yellow- — C. p. 

 passerina, C. p. jamaicensis, and C. p. albivitta. This agreement can 

 scarcely be considered indicative of close affinity in view of the present 

 distribution of these forms, which are isolated not only by wide water 

 areas, but also by the interposition of certain other forms quite 

 different in respect to the character in question. Again, the insular 

 race C. p. socorroensis finds its nearest relative, so far as color alone is 

 concerned, in a form from the interior of Colombia, C. p. parvula. 

 Aside from this, perhaps the most curious case is that of C. p. exigita, 

 a form which has developed on a small island between Porto Rico 

 and Haiti, but differs decidedly from the bird of either island, while 

 at the same time being exactly like the bird of Great Inagua in the 

 Bahama group. This of course may be a case of the re-duplication of 

 characters under similar conditions of environment. 



Enough has been said to show the difficulty of arriving at any 

 positive conclusions on this point. The writer's own views, so far as 

 they can be expressed in a linear sequence of names, are represented 



