97 
( ovember 26, 1869), state that in a set of specimens from the Pelew Islands, some had the rufous 
pectoral band, in two others it was entirely wanting, and in one bird there was only to be seen a faint trace 
of it ; ” and they therefore conclude that their so-called Rallus forsteri is nothing but a state of plumage 
due to age or season. The extent and colour of the facial band is likewise variable ; in some it is of a 
nch dark brown with well-defined edges, the grey plumage above forming a long narrow streak, while in 
others it is diffused, largely mixed with rufous, and spreading considerably on the hind neck. The distinct- 
ness of the white bars on the underparts varies in different individuals ; but this seems to be in some 
measure dependent on the age of the bird. An example which died in the Zoological Society’s Gardens, 
and was kindly forwarded to me by Dr. Sclater for examination, had the whole of the upper surface 
spotted with white, largely tinged on the wings with fulvous; others, again, I have seen in which the 
spotted markings were almost entirely confined to the hind neck and shoulders; but as it would be easy 
to bring together a complete intermediate series, this is of no value as a distinguishing feature. Mr. 
Potts’s so-called Rallus pictus, characterized by its decidedly superior size, would certainly be entitled to 
recognition but for the great variation in this respect to which this species is subject. The garter, or bare 
tibia, mentioned by Mr. Potts in his description of Rallus pictus (/. c.), is to be found also in' orfinary 
examples of our R. philippensis, although, of course, this feature is proportionally more conspicuous in the 
larger birds. No weight can be attached to the slight peculiarity in the shape of the bill, unless it should 
prove to be a constant character; for I can give an instance within my own experience of a very manifest 
modification in the bill of a Rail through purely accidental causes. On this point Dr. Pinsch writes to me as 
follows:— “I received in Ilaast’s last collection a specimen of the so-called Rallus pictus from the Okarita 
lagoon ; but I find that it differs in no way from those collected in the Pacific and elsewhere.” Mr. Gould 
also, in treating of this species *, regards the birds received from Southern and Western Australia, which 
are rather smaller and have more attenuated bills,” as mere local varieties. 
Among those in the Colonial Museum collection, one has the narrow snperciliary streak perfectly 
white in front ; another has the pectoral band of a rich buff colour, and about an inch in width, with the 
banded markings of the underparts very pronounced and extending up to the commencement of the fore neck 
Notwithstanding this extreme tendency to variation, I have never met with any instance of albinism 
except the one mentioned above. 
I have examined and compared a pretty extensive series from different regions with the following- 
result : — An example of Rallus philippensis from Fiji is more spotted than our bird, the round white spots 
spreading all over the mantle, wings, and upper tail-coverts; there is absolutely no pectoral band, not even 
an indication of it ; the nuchal collar of chestnut is very much enlarged, being about an inch and a half in 
depth, blending with the brown colour on the nape, but giving a rufous blush to the crown and forehead 
and extending in a broad bar across and somewhat under the eyes to the base of the upper mandible. In 
this tendency of the chestnut colouring to overrun the crown and vertex this bird shows an approach to 
Rallus striatus of India, in which the rufous crown and nape is a distinguishing feature. There is a further 
resemblance in the absence of the pectoral band ; but the striated character is entirely different, the wino-s of 
the latter being adorned with transverse and wavy lines or bars of white, which at once distinguishes this 
bird from all the others. In a bird from Pelew Island, on the other hand, there is only the slio-htest indi- 
cation of a nuchal collar, and the crown is faintly suffused with chestnut ; whilst a bar of dull chestnut 
brown covers the lores, passes through and under the eyes, and then becoming narrower, passes over the 
ears and fades away on the nape ; the spotted markings on the back and wings are less distinct, and in place 
of the pectoral band there is a mere wash of rufous yellow, forming a narrow zone. Moreover, the bill is 
decidedly more slender than in any of the preceding forms. The Australian bird is similarly marked to 
ours, the pectoral band and the banded markings on the underparts being very conspicuous, the former 
measuring more than half an inch in width, and being of a rich chestnut-brown. Owing to the absence of 
this interrupting pectoral band in the Fijian bird, the striped appearance of the underparts is very pronounced, 
especially as it reaches almost to the fore neek. In addition to this special feature, the bill, legs, and toes 
are appreciably stronger than in any of the other forms enumerated above. 
VOL. II. 
* Handbook to the Birds of Australia, vol. ii. p. 384. 
0 
