454 



BULLETIN 50, UNITED STATES NATIONAL MUSEUM. 



posterior ones), and usually (?) with more of brownish gray on 

 median portion of lateral rectrices." 



Adult male.— W\ng, 385-412 (402.8); taU, 112-133.5 (124.3); cui- 

 men, 74-83.5 (78.7); greatest depth of bill, 28.5-31 (29.8); tarsus, 

 33.5-37 (34.7) ; middle toe, 20.5-25 (23.3)." 



Adult female. — Wing, 382; tail, 119.5; culmen, 66; greatest depth of 

 bill, 24.5; tarsus, 28; middle toe, 17.^^ 



Seacoasts ^d larger inland waters of southern Brazil (Sao Paulo; 

 Iguape, Sao Paulo; Santa Catarina; Rio Araguaya; Espirito Santo), 

 Paraguay (Rio Pilcomayo; La Paz, Jan. 29; Gran Chaco), Uruguay, 

 and Argentina (Buenos Aires; Conchitas, Buenos Aires; Avellaneda; 

 Rio Parana); also, Pacific coast of Mexico and Central America (San 

 Bias, Tepic; Rio Zacatula, CoUma, April; Manzanillo, Colima, Feb. 11; 

 San Mateo, Oaxaca?; Acapam, Guatemala, Jan.).** 



oAn adult male from Paraguay, which in large size agrees closely with this form, 

 lias the under wing-coverts wholly pure white and has no more gray on the lateral 

 rectrices than some specimens of R. n. nigra. 



b Six specimens. 



c One specimen. 



Unfortunately none of the specimens of R. cinerascens and none of the South Ameri- 

 can examples of R. nigra intercedens have the sex determined; but each is readily 

 divisible into two series according to size, in all probability representing males and 

 females, respectively, the sexual difference in size being almost certainly as well- 

 marked in these two forms as in i?. nigra nigra. 



<^ Notwithstanding the anomalous distribution outlined above, I am unable to dis- 

 tinguish satisfactorily specimens from the Pacific coast of Mexico and Central America 

 (none have been examined, however, fro'cn the coast of Oaxaca) from true R. n. 

 intercedens. Most certainly they are not R. n. nigra nor R. cinerascens. Possibly a com- 

 parison of larger series from both the Atlantic coast of southern Brazil, etc., and Pacific 

 coast of Mexico and Central America may show differences sufficient to warrant the 

 Bubspecific separation of the latter. 



