262 Annals of the Transvaal Museum 
environment. This shows how necessary it is to exercise judgment in form- 
ing one’s conclusions, since either colour or structure may change to such 
an extent that it is easy to be misled ; to my mind, wide divergence from 
type indicates the conditions of environment and period of time under 
which the species have been under its influence. Our Serins, apart from 
these two genera, may be grouped as follows : Serinus Koch, type Fringilla 
serinus L., contains four subgroups in the species S. canicollis (Swainson), 
5 . scotops (Sundevall), S. mozambicus (Mull.) and S. angolensis (Gmelin). 
The hrst would seem to have radiated southwards together with Spinus 
from the Palaearctic region along the eastern escarpment, and to have 
become very distinct in the southern region, subsequently radiating north- 
wards again to near the Tropic of Capricorn. It differs from Serinus serinus 
in its relatively larger bill, shorter tail and style of coloration, the crown 
being golden yellow, underparts of the body yellow and grey on the sides 
of the neck. I propose to make it the type of a new subgenus, Pronospiza, 
type Serinus canicollis (Swainson). In the eastern escarpment forests is 
another species, which differs in being more greenish and heavily streaked 
below and the wing shorter, both characters due to the effect of environ- 
ment and sedentary habits; I propose to place this as another subgenus 
under the new name of Dendrospiza, type Serinus scotops (Sundevall). In 
the low country on the eastern tract from Kaffraria northwards, the place 
of these two is taken by S. mozambicus (Muller, cf. Iredale, Ibis, 1918, p. 
242), which I propose to place in another subgenus, Microserinus subg. 
nov., of which it will be the type; it differs from the two preceding in size 
and colour, from Pronospiza in its less forked tail and from Dendrospiza 
in its shorter tail. The fourth subgenus I propose to name Ochrospiza, 
type S. angolensis (Gmelin), which differs from the preceding in having 
yellow only on the rump, the underparts white with a black smudge on 
the breast and dark obscure streaks on the flanks ; in size it is still smaller 
than Microserimts, the bill is shorter, the tail shorter in proportion to the 
length of the wing and normally the primaries are longer; all these char- 
acters can be attributed to the dry conditions of its habitat. There is an 
overlap of these subgenera wherever the conditions of environment over- 
lap, and as their distribution extends over the continent we find species 
evolving, of which Alario would seem to be one, and if so must be regarded 
as only a subgenus of Serinus. 
Passing now to a plainly coloured genus in which yellow pigment is not 
developed, namely Poliospiza, type P. gularis (A. Smith), it would seem 
that this genus also originally found its way southwards along the eastern 
escarpment, but practically died out on the Drakensberg, leaving only 
P. leiicoptera Sharpe in the southern mountains; but with the change of 
conditions some individuals were able to survive by adapting themselves, 
of which we have evidence in P. gularis thait is widely dispersed to-day, 
andP. mennelli Chubb which has adapted itself to the fertile Lower Zambesi 
valley. As might be expected, these three species differ in structural char- 
acters, P. gularis having a narrow bill and longer wing, P. leucoptera has 
a heavier bill, generally darker colour and shorter wing, and P. mennelli 
has somewhat intermediate characters of structure and in colour stronger 
contrasts of black and white, such as is to be expected in the conditions of 
climate of the Lower Zambesi as compared with the southern Cape moun- 
