222 
Annals of the Transvaal Museum 
portion to the wing than in the preceding, the wing itself more rounded 
in shape. I make it the type of a new genus Notopicus. 
Thripias has lately been lumped with Mesopicus, but there is no sound 
reason why this should be done. 
C. Grant adopts the name of Dendropicos fuscescens Vieillot in pre- 
fer^^nce to guineensis and cardinalis, while D. hartlaubi is considered a sub- 
species of D. abyssinicus (cf. Ibis, 1915, p. 460). In the same place, Grant 
places Camp ether a malherbei as a synonym of C. cailliautii. C. capricornis 
of Damaraland is a very well-marked form, though possibly only a sub- 
species of bennetti. 
PiTTIDAE 
Finsch {Notes, Leyden Mus. xxiii. 206-212, 1903) has shown that 
Reichenow’s Pitta longipennis is identical with P. angolensis, which name 
should therefore be used for the species occurring within our limits. The 
Transvaal Museum possesses an immature specimen taken by P. A. Shep- 
pard at Beira, an adult which flew into a house at night in Pietersburg 
and an adult which was found dead under some tall trees in a garden at 
Potchefstroom. 
Hirundinidae 
Sharpe and Wyatt’s beautiful Monograph of the Swallows contains a 
reference to the importance of the nesting habits of the species for generic 
differentiation; but the genera were, nevertheless, not very carefully 
chosen, a number of well-marked genera being overlooked. In the genus 
Hirundo alone they lumped together widely different species that have 
little affinity to the typical ones. Hirundo rustica, the type species by 
general consent, has the nostrils opening through a narrow sublateral slit, 
the covering membrane being broad; the tips of the wings reach beyond 
the tips of the penultimate tail feathers and the forking of the tail (that 
is, taking the distance between the tips of these feathers and of the middle 
tail feathers, when folded together) greater than the length of the tarsus. 
Its nest is a shallow, open structure of mud lined with thin fibres and 
feathers and attached to some perpendicular rock or wall, close up under 
an overhanging ledge or roof ; its eggs are white, abundantly spotted with 
slate, brown and reddish markings. H. albigularis (Strickland), which 
breeds in South Africa, where .ff. rustica is a non-breeding migrant, differs 
from the typical species in that the wings do not extend beyond the tip 
of the “streamer” or outermost tail feather and it might, therefore, be 
subgen erically separated, if thought necessary. Another genus is to be 
found in H. filifera, genotype of Uromitris Bonaparte, and represented in 
South Africa by smithi, which has the forking of the tail to the tip of the 
penultimate feather not as deep as the length of the culmen, though the 
“streamer” is very long and narrow; its nest and eggs are similar to those 
of Hirundo, except in size. Another genus is to be found in H. atrocaerulea 
Sundevall, for which I propose the name of Natalornis, gen. nov., 
characterised by having the tail deeply forked (to the penultimate feather 
equal to the length of the tarsus, foot and first joint of the middle toe), 
the outermost tail feather in the male longer than the wing and the whole 
plumage blue-black in colour; its nest is of the same pattern as that of 
Hirundo, but the eggs have a more pearly gloss, though marked in much 
the same way. 
