MALURUS MELANOCEPHALUS: _Vig. & Horsf._

_J. & E. Gould del^t._ _C. Hullmandel Imp._

]

MALURUS MELANOCEPHALUS, _Vig. & Horsf._

Black-headed Wren.

_Scarlet-lacked Warbler_, Lewin, Birds of New Holl., pl. xiv.

_Malurus melanocephalus_, Vig. & Horsf. in Linn. Trans., vol. xv. p.

222.

_Malurus Brownii_, Jard. and Selb. Ill. Orn., vol. ii. pl. 72. fig. 1.

In their “Illustrations of Ornithology,” Sir William Jardine and Mr.

Selby have in a very laudable manner endeavoured to clear up what they

considered some confusion respecting the present and the preceding

species, _M. Brownii_. These gentlemen have, however, fallen into error

in considering the two birds as identical, whereas they are, in fact,

totally distinct.

I have never seen the Black-headed Wren from any other locality than New

South Wales, and I am consequently led to believe that the south-eastern

portion of Australia is its peculiar and limited habitat. It is a local

species, not being generally diffused over the face of the country, like

several other members of the group, but confined to grassy ravines and

gullies, particularly those that lead down from the mountain ranges. I

obtained several pairs of adult birds in very fine plumage in the

valleys under the Liverpool range, all of which I discovered among the

high grasses which there abound; but as the period of my visit was that

of their breeding-season, I never observed more than a pair together,

each pair being always stationed at some distance from the other, and in

such parts of the gullies as were studded with small clumps of scrubby

trees.

The Black-headed Wren has many actions in common with the _M. cyaneus_,

and like that species carries its tail erect: it also frequently perches

on a stem of the most prominent grasses, where it displays its

richly-coloured back, and pours forth its simple song. I did not succeed

in finding the nest, although I knew they were breeding around me: it

was probably placed among the grasses, but was so artfully concealed

that it completely baffled my efforts at finding it.

One might suppose the greater development of feather on the back of this

species to have been given it as a defence against the damp and dense

grasses of the ravines, among which it usually resides; but from the

circumstance of the female not possessing this character of plumage, and

the rich garb being only seasonal in the male, this supposition falls to

the ground. In their winter dress the sexes very nearly resemble each

other; but the males may always be distinguished by the black colouring

of the bill and tail-feathers. The young male of the year has the

tail-feathers brown, like the females, and it is a curious fact, that at

this age these feathers are much longer than in the adult.

The flight of this species is feeble and not protracted; but, on the

contrary, its powers of running and creeping are very considerable.

The breeding-season probably commences in September and continues until

January; its food is insects of various kinds.

The male has the head, all the under surface, wing-coverts, upper

tail-coverts and tail deep velvety black; back of the neck, scapularies

and remainder of the upper surface rich orange-scarlet; bill black; eyes

blackish brown; feet fleshy brown.

Female brown above, paler beneath; bill brown; base of the under

mandible reddish brown; feet flesh-brown.

The Plate represents male and female in the summer plumage, and a young

male in change, on one of the native grasses of New South Wales.

[Illustration:

The Birds of Australia, Vol. 3 of 7 · 第 53 / 85 题