Observations on the Cubital Coverts of the Huornithes. 179 
Anseres, Steganopodes (Cormorants and Pelicans only), 
Fulicarie (exclusive of Heliornis), Pygopodes, Colymbi, and 
the Guillemots. In the whole of these the 5th cr. is 
functionally absent, and consequently the Majors, Medians, 
and Minors are, all alike, faulted over the place of the 
defunct remex. The first six Medians, and usually not less 
than three following, and their three or four rows of cor- 
responding Minors, show proximal overlap. The swmber 
of feathers in each row appears to be directly related to the 
length of the cubitus. Where this is comparatively short, 
the abbreviation of the feathers takes place at the expense 
of such coverts as lie nearest to the cubito-humeral joint, and 
at the same time overlap proximally. In other words, as the 
forearm is shortened, in adaptation to changed modes of life, the 
covert feathers first to disappear are not those directly at the 
proximal end of the cubitus, but only such of them as show 
proximal overlap—those terminal feathers that show distal 
overlap apparently in all cases surviving the change, and 
advancing to a position nearer and nearer to the carpus. It 
is not easy to say why this should be so; but the facts seem 
to point to no other conclusion. Conversely, with increased 
length of forearm, the covert feathers added on come 
between those showing proximal overlap and those showing 
distal overlap, which invariably form the end of the series. 
This consideration will enable the reader to perceive the 
distinction between the Rails (exclusive of Helzornis), the 
Gallinules and Coots, the Grebes (whose wing style marks 
them as Diving Rails), and the Colymbide. Whether the 
ancestral form of the Landrail was Colymboid, or whether 
Colymbus is a marine Grebe, and, therefore, a Rail, whose 
structure has been adapted to a mode of life in the open sea, 
cannot be definitely stated. If I might hazard a guess, it 
would be to the effect that Colymbus is the nearest living 
representative of the ancestral form; that the Crakes represent 
one section of its descendants modified to a life inland; and 
that, possibly, the Grebes are reverting from the Ralline 
stage of modification to one more nearly resembling Colymbus. 
Heliornis appears to be a quincubital Rail, judging by a 
specimen lent to me by Mr Beddard. The relationship of 
