that he never heard it utter a note. He also met with it during the cold season in the low forest-
regions of Singbhoom in Western Bengal, where he appears to have had a better opportunity of
watching its habits. Although it prefers swampy ground, such as is found covered over with thick
bushes near deserted tanks, or marshy ravines in the jungle overgrown with bamboos and underwood,
it is nevertheless also found in dry stony Saul jungle. In its habits it appears to resemble its
congeners. It perches low and keeps much to the ground, hunting amongst the dead leaves for
earthworms, slugs, leeches, &c. When approached, it flies off swiftly, keeping near the ground, and
seeks some dense covert, where it remains motionless and silent. The birds are generally found
singly» but sometimes three or four may be seen together near the thickets and dense underwood,
and, when disturbed, they fly off to different places of concealment.
During the cold season, when it lives in the plains, it occasionally finds its way to the
gardens, where it is recorded as very wary, feeding under the bushes, and, when disturbed, flying up
into the mango-trees (Beavan, Ibis, 1870, p. 827).
I t is probably a partial migrant on all the hill-ranges within the area of its distribution,
descending to the adjoining plains in the cold season (Oates, Fauna of British India,. Birds, ii.
p. 152).
Sir V. Ball invariably found it on the banks of well-wooded streams, and remarks that its flight
was low and irregular. Mr. Oates says that it feeds on the ground, and Jerdon makes the same
remark, adding that it devours fruit and seeds. It is said to have a pleasant song, not unlike that of
Turdulus wardi (G. F. L. Marshall, in Oates’s edition of Hume’s ‘ Nests and Eggs of Indian Birds,*
ii. p. 107j1.
It breeds in May and June, laying two or three eggs in a nest built rather substantially of moss,
neatly lined with stalks of maiden-hair fern and a few bents of grass. The nest has been found in
the fork of a moss-covered rhododendron, twenty feet from the ground (G. F. L. Marshall).
The eggs are very similar to those of the Common Blackbird, and vary in colour, size, and shape
to the same extent. An egg in the British Museum, obtained by Mr. W. E. Brooks at Gulmurg in
Kashmir on the 6th of June, 1871, measures 1*28 by *91 inch. The pale greenish-blue groundcolour
is almost entirely hidden by the reddish-brown markings (Brooks, Stray Feathers, iii. p. 237).
Geocichla dauma exactly resembles G. varia in all the various details of its plumage so far as
colour is concerned, but it differs from its Siberian ally in the following particulars:—
Instead of four, there are only three very dark feathers on each side of the four paler brown
feathers in the centre of the tail, so that the total number of rectrices, instead of being 14, is only 12.
Instead of having the second primary intermediate in length between the fourth and fifth, this is
intermediate between the fifth and sixth, so that the wing is slightly more rounded, as might be
expected in a species whose range of migration is so much more restricted.
It is also, on an average, a smaller bird than.O. varia. Length of wing 6*0 to 5v3 inches, tail
4*4 to 3 6 inches, culmen IT to TO inch, tarsus 1 3 2 to 1*2 inch. Outer tail-feathers 0*3 inch shorter
than the longest. First primary shorter than the primary-coverts.
In what I take to be the bird of the year, the ground-colour is much more ochraceous than in
the adults, and the bastard-primary is almost as long as the longest primary-covert. In abraded
summer plumage the ground-colour is much less ochraceous. In a very young bird in the British
Museum most of the first plumage is retained; the black lunulations are almost obsolete on the
rump and upper tail-coverts; the under tail-coverts and the breast are very buff, but the bastard-
primary is shorter than the primary-coverts.
The amount of white at the tips of the tail-feathers varies more than usual in this species, in
some examples extending for nearly an inch on the inner web of the outer feathers, and in others
being almost obsolete.
The Geocichline patches on the under surface of the wing are nearly always buff, but there are
two examples in the British Museum in which they are white: it is probable that these are in
the fully adult plumage. So far as is known, these Geocichline patches are always white in
Geocichla nilgirienm and G. heinii, and nearly always buff in the other species of the sub-genus
Oreocincla.
There is an example in the British Museum, dated the 27th of September, in which some of the
quills are still “ in the pen.”
The specific characters of Geocichla dauma may be summed up as follows:—
I. General colour of upper parts olive-brown, and of underparts white (more or less suffused
both above and below with golden buff in autumn, especially in the first autumn); nearly all
the body-feathers, both above and below, ornamented with black crescentic tips [distinguishing it from
every species belonging to the genera and sub-genera Zoothera, Psophocichla, Cichlopasser, Turdulus,
and Geocichla, but not from those belonging to the sub-genus Oreocincla].
II. Rectrices twelve in number [distinguishing it from G. varia and G. horsfieldi].
III. Rump and upper tail-coverts, in summer, olive-brown (not russet-brown), lunulated with
black [distinguishing it at that season from G. horsfieldi, G. nilgiriensis, G. papuensis, G. heinii, and
G. macrorhyncha].. .
IV. Feathers of crown and nape crossed with very conspicuous pale sub-terminal bars varying
in colour from nearly white in adult summer p lumage to golden buff in first autumn dress
[distinguishing it from every other species of the sub-genus Oreocincla, except G. varia
V. White terminal patch on inner web of outermost tail-feathers generally less than half an inch
long [generally distinguishing it from G. heinii, G. cuneata, and G. papuensis].
VI. Tail moderate in length, less than four-fifths of the length of the wing [distinguishing it
from G. macrorhyncha, G. Iv/nulata, and G. cuneata].
VTI. Bill comparatively small, the culmen less than a third of the length of the tail
[distinguishing it from adult examples of G. nilgiriensis and G. imbricatd].
The following combinations are diagnostic within the sub-genus:—I., II., IV .; I., II., III.,
VI., VII.; I., II., V., VI., VII.
A figure of this species has apparently never been published before, but there are various
sketches in the Hodgson portfolios in the British Museum (Gray, Catalogue of the Specimens and
Drawings of Mammalia and Birds of Nepal and Thibet presented by B H. Hodgson, Esq., p. 80).
In the first edition of that ‘ Catalogue,’ published in 1846, Gray called the bird Turdus whitei, but in
the second edition, published in 1863, the name is corrected to Oreocincla dauma.
The right-hand figure in the Plate is a life-sized representation of an example in my museum
procured in 1877 in Sikhim by the collectors of that indefatigable ornithologist, the late Mr. L.
Mandelli, of Darjeeling. The smaller figure in the distance, to the left, is taken from a skin shot by
Mr. A. Anderson on the 5th of May, 1895, in Xumaon; it is a male in the somewhat faded summer
plumage, which has been described as Oreocincla parvirostris.