272 



Murphy, Anatidm of South Georgia, 



rAuk 

 Ljuly 



from frontal feathers, 32-36; width of bill at base, 12.5-16; tarsus, 

 35.5-39; middle toe and claw, 45-51. 



Four females, collected between December 1 and March 3. 

 Length (skins), 390-412; wing, 195-207; tail, 85-93; culmen, 

 from frontal feathers, 31-34; width of bill at base, 12-15; tarsus, 

 35-36; middle toe and claw, 46-49. 



The testes of a male shot on December 1, 1912 measured 38 X 19 

 millimeters. 



The crop of a female collected January 2, 1913, contained marine 

 amphipods. 



Salvador! (/. c, p. 264), without having seen a specimen of 

 Nettion georgicum, concludes that its affinities are with the group of 

 teals containing the South American species A'^. flamrostrc, N. 

 oxypteruvi, and N . aridium. A comparison of my specimens with 

 all of these, however, shows that the South Georgia bird is quite 

 distinct. Its real relationship, hitherto unsuspected, is with the 

 duck known as Dafila spmicauda (Vieill.), a widely distributed 

 species, occurring, apparently in the form of several undescribed 

 geographic races, from Brazil to the Straits of Magellan and the 

 Falkland Islands. The South Georgian teal is, indeed, almost a 

 facsimile of Dafila spinicauda, smaller, considerably darker (espe- 

 cially on the under surface), but with similar proportions, the same 

 pattern and distribution of color over the whole body including the 

 bill (fide R. H. Beck, label), the same wholly black speculum with a 

 green sheen, the same black-striped tertials and pointed tail. Dr. 

 Frank M. Chapman, who first called my attention to the striking 

 resemblance between the South Georgia birds and skins of Dafila 

 spinicauda in the magnificent Brewster-Sanford collection, remarked 

 at the same time that the case furnished an excellent example of 

 taxonomic relationship obscured by inaccurate nomenclature. 



