INTRODUCTION. xvii 



" I cannot omit taking notice of a small bird which appears to be an annual visitor to this 

 island, as they have been here about the same time both last year and the present one. They 

 come in immense flocks, fly rather high and in waves. They are evidently a seed-bird, of the 

 Sparrow kind, and very much resemble the wild Canary both in colour and size. They only 

 remained here a few days ; and I fancy they went away on the 3rd of this month [April], which 

 was a fine day with a light southerly breeze." 



The genus Zoster ops comprises a rather numerous group of closely allied species, with a wide geographic 

 distribution; but, as a genus, it is somewhat isolated in its affinities. Mr. Gould in placing it, very properly, 

 next to the Australian Honey-eaters, observes : — " I have been influenced by their approximation to these birds 

 both in form and habits, and to which they exhibit a further degree of affinity in the form and structure of 

 their nests, but not in the colouring of their eggs, which are always blue." But I would remark that a 

 stronger indication of this affinity than any mentioned by Mr. Gould is to be found in the structure of the 

 tongue, which is slightly pencilled at the tip, and proclaims at once the meliphagous habits of the group. 



Members of this genus are scattered through Southern Africa, India, China, and Japan ; but the species 

 are most numerous in the sea-girt lands of Australasia and Polynesia, where each group of islands appears to 

 have one or more species peculiar to itself. Mr. Gould records three well-marked species from Australia, two 

 from Lord Howe's Island, and two more from Norfolk and Phillip Islands. There is one species (Zosterops 

 flamceps) peculiar to the Fiji Islands, another (Z. flavifrons) to the New Hebrides group, and another (Z. con- 

 spicillata) to the Ladrone or Marian Islands. Two species inhabit New Caledonia (Z. xanthrochroa and 

 Z. griseonota) ; one (Z. cinerea) is recorded from the Caroline group, and another [Z. melanops) from the 

 Loyalty Islands. 



The New-Zealand representative of the genus is undoubtedly the same as Zosterops lateralis, Lath. 

 { = ccei~ulescens, Gould), an inhabitant also of Tasmania, New South Wales, and South Australia. 



Fam. Stuemd^e. Naturalists who profess to be governed by the ordinary rules of zoological 

 nomenclature will, I feel persuaded, follow me in the adoption of Eeteralocha acutirostris in 

 substitution for H. gouldi, as applied to the Huia. But it is necessary to alter the position of 

 the bird in our system of classification, inasmuch as it proves, on further research, to be a 

 Passerine form, and not a Picarian. In placing it in the family Upwpidce (order Picarise) I 

 stated (at p. 63) that till its affinities were better understood I preferred to leave it where my 

 predecessors had stationed it, especially as I had myself observed a striking similarity in some 

 of its habits to those of the Common Hoopoe {Upupa epops). The bird which I referred to 

 at page 68, as then living in the Zoological Society's Gardens, has since died ; and the loss to 

 the collection has, in this instance, been the gain of science ; for Mr. A. H. Garrod has thus 

 been afforded an opportunity of studying the osteology and anatomy of this singular form, the 

 result being that he assigns it a place among the Starlings. It is only fair, however, to mention 

 that the late Mr. G. E. Gray, in his ' Hand-list of Birds,' had already referred it to the family 

 Sturnidce. ■ Here it still holds an isolated position as the only known representative of the genus, 

 although it seems to have a near generic ally in another New-Zealand form, the Creadion carun- 

 culatus or Saddle-back. 



I quote the following from Mr. Garrod's valuable paper read before the Zoological Society 

 on the 21st May last : — 



" The arrangement of the feathers is completely Passerine. The rhombic saddle of the spinal tract does 



c 



