Annals of the Transvaal Museum 209 
naming the genera they represent by Sclater {Bull. Brit. Orn. Cl. XLii. 75, 
1922), hellicosus being referred to Pblemaetus Heine and coronatus to a new 
genus Stephanoaetus. 
The genus Circaetus Vieillot, like Circus, contains migratory and resident 
species and should be divided into several subgenera: Circaetus fasciolatus 
Gray and C. cinerascens Vieillot might for the present be retained in the 
typical genus, as they appear to me to be resident representatives on the 
east and west, respectively, of the genotype, C. gallicus, of the north. 
C. cinerascens, which is recorded from our limits by Reichenow (cf. Vog. 
Afrikas, i. 573, 1901), is larger than C . fasciolatus , but the latter appears 
to increase in size northwards, so that this difference cannot be regarded 
as of generic value in this case. C. pectoralis A. Smith differs so much in 
plumage characters that I propose to place it in a new subgenus under the 
name of Smithaetus; and finally C. cinereus Vieil. is again so markedly 
different in colour characters that there is no option but to separate it as 
well, under a new subgeneric name, for which I propose Melanaetus. 
In regard to my record of C. gallicus from South Africa {Ann. Transvaal 
Mus. IV. 109, 1911), the specimen would appear to be referable to the 
young of C. pectoralis, as stated by Finch-Davies (cf. Ihis, 1920, p. 419). 
Gurney {Ihis, 1876, p. 467) long ago pointed out that Buteo augur and 
B. jakal (= rufofuscus Forster, cf. Kirke-Schwann, Handlist of the Birds 
of Prey, 1920) should be referred to the genus Pterolestes, and this course 
should be followed. The name of Buteo desertorum (Daudin) should never 
have been used for our bird, Le Vaillant’s plate, on which it is founded, 
being quite unrecognisable; Sclater {This, 1919, p. 254) has recently adopted 
the name of B. rufiv enter ■ ]ei 6 .or\ on that account, following Hartert’s 
rejection of the name of desertorum {Vog. Pal. F. 1914, p. 1125) in favour 
of B. anceps Brehm. There is, however, still an element of doubt as to this 
species, some individuals being known to breed in the Cape Province, 
though the greater number found within our limits are undoubtedly 
migrants. 
Recent authors do not recognise Milvus migrans (= korschun auct.) as 
a migrant to South Africa. M. aegyptius should be known as M. aegyptius 
parasitus (Daudin) (cf. Hartert, Bull. Brit. Orn. Cl. xxxiii. 90, 1914) the 
typical form being found only in the far north-east. 
Helotarsus ecaudatus = Terathopius ecaudatus (cf. Richmond, Auk, 1902, p. 92). 
Baza verreauxi = Aviceda verreauxi (cf. Mathews, B. Australia, v. 212, 1915). 
The classification of the Falcons and Kestrels is not satisfactory, 
authors vainly attempting to fit the species into genera with which they 
have little in common, and by so doing hopelessly confusing their phytogeny. 
It is evident from the number of generic names applied during the last 
century that the authors of that period were observant and dissatisfied 
with lumping methods, and we have therefore now, by returning to a 
system of niore natural arrangement of the groups, generic names for nearly 
all of them. Throughout the Accipitres we find that colour characters are 
remarkably prominent in indicating the phylogeny of the species, structural 
characters often changing to so great an extent that if we are guided by 
them alone problems of phylogeny are most difficult to settle. The differences 
between banding and striping of the plumage, for example, remain so 
