[[p. 18]] This collection of birds was made by myself during two months of the year 1861. It consists of sixty-six species, among which were no less than seventeen new ones. Of these, three were found about the same time in the Island of Sula, and, with a new Pitta, have already been described in the Society's 'Proceedings,' leaving thirteen to be described in the present paper.

     In my paper "On the Birds of the Sula Islands," read before the Society at their last Meeting, I pointed out that the large proportion of purely Celebes forms found there forced us to the conclusion that a closer connexion had once existed between those islands and Celebes, and required us to class them as forming a single zoological group. The Island of Bouru must, on the contrary, be classed with the Moluccas; for leaving out about twenty species of rather wide distribution, the remaining forty-six are all either identical with, or most nearly allied to, Moluccan species. Not a single characteristic Celebes form is found in Bouru; and there are only three birds in the island whose affinities seem rather with the Indian than the Australian region, viz. Alcedo moluccensis, Hirundo javanica, and Treron aromatica.

     Bouru is therefore the western limit of the Moluccan fauna, and is the poorest portion of it, having several very remarkable deficiencies. Lorius, found in every other island of the Moluccas and New Guinea, is absent; Cacatua, found in every island of the Australian region, is also absent; and, stranger still, Buceros and Corvus, found in almost every large island of the archipelago, are both wanting. [[p. 19]] With these exceptions, most of the Moluccan types are represented either by identical or allied species.

     The following is a list of the new species now described, and of a few others which seem confined to Bouru:--

Tanygnathus affinis, n.s., Bouru and Ceram.
Accipiter rubricollis, n.s., Bouru and Ceram.
Athene hantu, n.s.
Tanysiptera acis, n.s.
Ceyx cajeli, n.s.
Pitta rubrinucha, Wallace.
Cisticola rustica, n.s.
Mimeta bouruensis (H. & J.).
Criniger mysticalis, n.s.
Monarcha loricata, n.s.
Rhipidura bouruensis, n.s.
Tropidorhynchus bouruensis, n.s.
Campephaga marginata, n.s.
Dicæum erythrothorax, Less.
Nectarinea proserpina, n.s.
Gallinula frontata, n.s.

     All but the two first species in this list are confined to Bouru only, and they are mostly representative species of Moluccan forms. Besides these, the three species of Pachycephala are also, as far as the Moluccas are concerned, peculiar to Bouru; for though they are found also in Sula, they have evidently emigrated there, the Celebes group, to which Sula belongs, not possessing any species of the genus. The Island of Bouru may therefore be considered to have added seventeen new species, but not any new forms or genera, to the Moluccan avifauna . . .

     [[p. 26]] . . . Remarks.--This curious bird [[Mimeta bouruensis]] resembles so closely a Honeysucker of the genus Tropidorhynchus that it has been figured and described as such, and even escaped the acute eye of Prince Bonaparte, who has given it that place in his 'Conspectus.' But, more singular still, there is a species of true Tropidorhynchus inhabiting the same island of Bouru, which so closely resembles this bird that the two can hardly be distinguished, except by a close comparison of the generic characters that separate them. We have here, in fact, a case among birds of that mimicry of one species by another belonging to a different group, which Mr. Bates has so well illustrated among the Lepidoptera of S. America (see Linn. Trans. vol. xxiii. p. 495). In this case the Oriole has imitated the Honeysucker; for it has [[p. 27]] departed from the usual gay colouring of its allies, and is actually the dullest-coloured of its family, while the Honeysucker very much resembles in its coloration other species of the group to which it belongs. The imitation is carried to the minutest particulars: the bare black orbits of the Tropidorhynchus are copied by a patch of dusky feathers in the Mimeta; the rigid lanceolate feathers on the head of the former are imitated by dark stripes on the broader feathers of the latter; and even the very peculiar ruff of recurved feathers on the nape of the Tropidorhynchus has its general effect imitated by a collar of a pale colour in the Mimeta. The under and upper surfaces of the two birds are as near as possible of the same tint respectively; and, stranger still, the Oriole has closely copied the mode of flight and the voice of its model; so that in a state of nature the two birds are practically undistinguishable. Most of the species of Tropidorhynchus have an elevated keel or protuberance at the base of the bill. In the Bouru bird this is altogether wanting; yet in the Mimeta which copies it there is a slight protuberance at the base of the bill, which does not occur in any other species of its genus--almost making us think that some ancestors of the present bird had mimicked a species of Tropidorhynchus which possessed the protuberance, and that their descendant, finding himself in the company of a bird without this ornament, was gradually losing it, but had not yet quite done so. It has been observed by Mr. Bates, and is no doubt generally true, that mimicking species are much less abundant than those they copy. In the present instance it seems to be different; for I obtained many specimens of the Mimeta before I saw a single Tropidorhynchus, though in other islands the latter was generally the most abundant. Perhaps in this case it has carried the imitation to such an extent as actually to gain an advantage over its model in the struggle for existence. This curious instance of mimicry does not stand alone; for in the adjacent island of Ceram, two allied but very distinct species (Mimeta forsteni and Tropidorhynchus subcornutus) resemble each other with equal accuracy. What peculiar immunity from danger the Tropidorhynchi possess, which makes it advantageous for other birds to imitate them, it is not very easy to see. In the case of insects, it seems probable that it is the odour or taste of the imitated species which is unpalatable to insect-eating birds; or, in other cases, like the clear-winged Moths which mimic Hymenoptera, the mimicked species are armed with a sting. In birds it is evident that the bravest, strongest, and best-armed groups should be the subjects of mimicry, and the weakest and most defenceless those which obtain some advantage by imitating them. Now this is certainly the case, for the Raptores are the most frequent subjects of imitation--a Parrot (Strigops) imitating an Owl, some Curassows of the genus Ibycter resembling Hawks (Ibis, vol. ii. p. 223), and Cuckoos frequently resembling Hawks. A species was named by Temminck Falco cuculoides; and in all parts of the world the larger grey and banded Cuckoos are mistaken by the natives for Hawks. Cuckoos, however, which are certainly among the weakest and most defenceless of birds, imitate several [[p. 28]] other groups, especially Gallinaceæ,--for example, Centropus phasianus in Australia, and Carpococcyx radiatus in Borneo, which latter is terrestrial in its habits, and much resembles the Euplocami of the same island. Eudynamis also frequently resembles Pigeons, especially the females and young birds, which are banded like Macropygia. Among the small Cuculinæ some are very like Campephagæ; and Chrysococcyx has put on the metallic plumage of Lamprotornis.

     Returning now to Mimeta and Tropidorhynchus, we have to observe that the former is a smaller, weaker, less active, less noisy, and less pugnacious bird; the feet have a less powerful grasp, and the bill is less acute. The latter has a great variety of loud and piercing notes, which bring its companions to the rescue in time of danger; and I have observed them drive away crows and even hawks which had ventured to perch on a tree where two or three of them were feeding. The Tropidorhynchus knows how to take care of himself, and make himself both respected and feared; it would therefore evidently be to the advantage of the more defenceless Mimeta to be mistaken for him.

     In this instance, as in most others, the imitation is far closer in the living bird than in the dead specimens, and it is a far more satisfactory case of mimicry than any of those which I have alluded to as occurring among birds, and which are more or less general resemblances to another group; while here we have two species, each confined to a single island, and each accurately imitated by a bird of a distinct family, with which it has no direct affinities.

     I therefore cannot doubt that this is a true case of mimicry, exactly analogous to that so common among insects, and which my friend Mr. Bates has the honour of having first brought under the same general laws which have regulated all variation in the organic world.

Note Appearing in the Original Work (at the bottom of pages 21-22)

*Since reading this paper [[i.e., G. R. Gray, P. Z. S. 1860, p. 344]], I have seen Professor Schlegel's 'Catalogue of the Birds in the Leyden Museum,' part 1, in which (Astures, p. 39) he describes this bird [[Accipiter rubricollis]] under the name of Nisus cirrhocephalus ceramensis, which seems to be equivalent to making it a variety of N. cirrhocephalus. Considering, however, the [[p. 22]] bird to be a very good species, I should at once have adopted Professor Schlegel's name ceramensis, had I not obtained the bird in other localities than Ceram. The Raptores having so generally an extensive range renders the application of territorial specific names less advisable in their case than in that of the Passeres. My own rule is only to apply the name of a country as specific name when the surrounding districts are known to possess their peculiar representative species, in which case it amounts almost to a certainty that the new bird is similarly restricted in range.