f)00 BIRDS OF PARADISE AND BOWER BIRDS STEJNEGER. 



ones overlap distally while six posterior ones overlap proximally, and, 

 moreover, that there is above them a well differentiated series of minor 

 coverts (as distinguished from the marginals) the overlap of which is 

 distal; in other words: The arrangement is absolutely in every particu- 

 lar as described by Mr. Goodchild for the Corvidoe. If any one will 

 take the trouble to examine Mr. Goodchild's own figure representing 

 the cubitals of a Bird of Paradise (Fig. 3), he will find that in the row 

 of medians there are two feathers svith proximal overlap, thus differing 

 materially from his description. 



It is plain from the above that the Paradisceince proper form no excep- 

 tion from the genuine passerine arrangement of the cubitals. 



Mr. Goodchild makes out another exception,* however, for later on 

 he states that the Bower Birds (which I regard as a subfamily of the 

 Paradisceidw) have a picarian arrangement of the minor coverts: " In 

 these more than one row of minor coverts occur, each with proximal 

 overlap." He even goes so far as to admit as a possibility that subse- 

 quent research may prove that the birds in question rightly belong to 

 the Picarians and not to the Passeres at all (not to speak of his class- 

 ing the Bower Birds with the Cotingidce among the Mesomyodi). 



With a splendid male of Ptilonorhynchus violaceus (U. S. Nat. Mus. 

 No. 121196) before me, I must again disagree with Mr. Goodchild's 

 conclusions. The firm structure of these feathers, coupled with the 

 metallic gloss of the margins, makes it easy for us to observe that a 

 few of the anterior medians have distal overlap, while at least eight of 

 the same series overlap proximally, and that the first row of minors 

 overlap distally as in a well regulated passerine bird they ought to do. 



There can, therefore, be but little doubt that the Bower Birds in 

 these respects conform with the Paradise Birds and with the Corvidce. 

 Mr. Goodchild's "Tabular View," on page 331, is thus fortunately purged 

 of a feature which might be urged as a reason against the view held by 

 him of the great value of the cubital arrangement for classificatory 

 purposes, a view which I am happy to say that I have held ever since 

 Igav^e ornithological classification any thought, but of which I became 

 somewhat doubtful when I saw the Paradismdcv separated from the 

 other Passeres and included in a group with the Hummingbirds, 

 Swifts, and Trogons. 



'This exception is not noted in the scheme on p. 331, 



