Annals of the Transvaal Museum 
231 
‘‘Tractrac” {Ois. d' Afriq. PL 184 and text), on which Vieillot based the 
name of Oenanthe cinerea which is illegally used by modern writers. Sharpe 
could not have critically examined Le Vaillant's work, either when he 
described Saxicola layardi (Birds of South Africa, p. 236, 1877) nearly 
thirty years later (Ibis, 1904, p. 325), or he would have perceived that 
Vieillot’s cinerea and his own layardi were synonyms. In 1904, Sharpe 
gave a synopsis of the species pollux, cinerea and schlegeli, stating that the 
first had a grey, the second and third a white rump, cinerea differing from 
schlegeli by its larger size. Le Vaillant compared his “Tractrac” with the 
Familiar Chat (Phoenicurus familiar is) and both the plate and description 
agree quite well with the characters of Sharpe’s Saxicola layardi; and, 
moreover, C. H. B. Grant (Ibis, 1911, p. 413) and other naturalists have 
noted how very similar are the two species, so that Le Vaillant’s comparison 
was a natural one. The type locality for the “Tractrac” is Outeniqua and 
it would seem to occur thence northwards to the Orange River, overlapping 
the range of Emarginata sinuata and Phoenicurus familianis, so that it is 
to be regarded as a firmly established species, and as such a genus, for 
which I propose the name of Phoenicuroides. I have not yet seen a 
specimen from Damaraland or Great Namaqualand which could be referred 
to either schlegeli or what authors have recently regarded as cinerea; but 
if the difference in size mentioned by Sharpe is not a matter of sex, the 
Great Namaqualand bird will have to be named. No doubt these two 
species are referable to Karrucincla. 
With regard to the name of Oenanthe cinerea Vieillot, recent authors 
have ignored the fact that the name was first used by Vieillot (in the same 
publication) for 0 . penairthe (L.), and that it was consequently invalidated. 
Mathews and Iredale (Austral. Av. Rec. iv. 144, 1921) have recently pointed 
out that there is an earlier name for the “Tractrac” in Motacilla tractrac 
Wilkes (1817), which therefore becomes available for the genotype of 
Phoenicurpides. We have thus three species which I here place in as many 
genera, the first Phoenicurus familiaris, which is widely dispersed and 
perhaps the ancestral form, with the second primary without a sign of 
emargination, the second Phoenicuroides tractrac with a local distribution 
in the south-west, having the second primary slightly emarginate, and 
Emarginata sinuata with a rather wider distribution in South Africa, 
extending to Waterberg, Transvaal, and having the second primary very 
deeply emarginate. The larger species with the second primary emarginate, 
Karrucincla pollux, appears to me to have had a different origin and to 
be more akin to Grillivora. 
Yet another chat requires generic separation, namely Saxicola albicans 
Wahlberg, which is characterised by its much stouter bill, legs and feet 
as compared with Phoenicurus familiaris, which occurs side by side with 
it ; it differs from Emarginata, Phoenicuroides and Karrucincla in showing 
no sign of emargination on the second primary and it differs also from 
Grillivora in coloration, shape of wing and smaller size. I, therefore, 
allocate it in a new genus under the name of Psammocincla. 
Saxicola arnotti and its allies are more nearly related to Myrmecocichla 
than Thamnolaea, with which it is sometimes associated; but it differs from 
Myrmecocichla in its more rounded wing and absence of white on the under 
surface of the primaries. It is a forest bird and differs accordingly from 
