182 Proceedings of the Biological Society of Washington. 



Massachusetts, for similar courtesies. We are also indebted 

 to Drs. Charles W. Richmond and Harry C. Oberholser for their 

 kindness in verifying certain references. The present paper is 

 governed by the same conditions as the first of the series. 



The South American forms of Myiarchus fall naturally into 

 four groups, typified respectively by M. tyrannulus, M. pelzelni, 

 M. ferox, and M. tuberculifer. They range from Panama south- 

 ward to Argentina, throughout the Tropical Zone, while all of 

 the groups except the first have a single representative each in 

 the Subtropical. One species is a winter visitant from eastern 

 North America. M. pelzelni is said to be a bird of the campos 

 region in Brazil, but the species in the main are forest-dwellers, 

 with typical flycatcher habits, and nesting in holes in trees, in 

 the well-known manner of M. crinitus. Wear and fading affect 

 the color of the plumage greatly, while birds in ju venal and 

 immature dress show more or less decided rufescent edgings on 

 the wings and tail. Great care in making comparisons for 

 geographical differences is thus indicated, while another source 

 of error to be guarded against arises from the circumstance that 

 preserved skins fade out with age, old specimens being appreci- 

 ably paler than those freshly collected. The diagnoses and 

 descriptions in the present paper are based (unless otherwise 

 specified) on adult specimens in fresh unfaded plumage. 



References to the birds of this group are very numerous in the 

 literature of South American ornithology, but the various forms 

 have been so imperfectly understood that the synonymy is 

 exceedingly involved. In many cases it is difficult or impossible 

 to determine from a study of the text what form or forms cer- 

 tain authors had in hand, and it is to be hoped that those 

 European ornithologists who may have access to the specimens 

 (if extant) on which such records were based will look them up 

 and publish their results. The first serious attempt to review 

 the group as a whole was by Coues in 1872. His treatment is 

 very unsatisfactory, partly because of insufficient and poor 

 material, and partly by reason of failure to reahze the signifi- 

 cance of the characters in this group. In some respects he even 

 left the nomenclature in a more confused state than before. 

 Sclater's treatment in Volume XIV of the Catalogue of the Birds 

 in the British Museum (1888) was a great improvement, 

 although he did not recognize subspecies. Mr. Ridgway indi- 



