294 Messrs. Mathews and Iredale on a 



criticism of the birds preserved would simply convince the 

 observer o£ the incongruity of such a conclusion. The lack 

 of a series, however^ negatived any subdivision that would gain 

 acceptance^ and a study of the literature revealed a similar 

 hesitancy upon the part of previous workers, most fully 

 acknowledging the incompleteness of their conclusions. 



First, it should be observed that there has been little 

 confusion regarding the North Island Wood-Hen. Though 

 comparatively quite a number of birds are in collections 

 from that Island, these show little variation, and there has 

 been no confusion in its nomenclature. The climate and 

 topography of that Island does not show the variation that 

 is found in the South Island. 



The first striking form met with in the South Island is 

 that known as the Black Wood-Hen. We admitted this 

 form to specific rank as it occurs alongside a form of 

 normal coloration. Our conclusion was that it was a fixed 

 melanistic form breeding true, and confined to the wet^ 

 heavy-bushed south-west coast of the South Island, It is 

 replaced on the tops of the adjoining mountains by a pallid 

 bird, to which Hutton gave the name Ocydromus hectori. 

 The lowland form found in company with the Black Wood- 

 Hen was first described as Rallus australis Sparrman. It has 

 puzzled every investigator to decide whether these three are 

 specifically distinct or not, and much of this difficulty is due 

 to lack of specimens. In our Reference List we decided in 

 favour of three species, but we could not delimit the sub- 

 species into which these species were certainly divisible. 

 However, as the name G. hectori (Hutton) was given to a 

 n^ouiitain bird, it seemed certain that it was inapplicable 

 to the common lowland form^ as every writer had noted 

 previously. 



The examination of the Reischek collection in Vienna 

 by one of us has proved that the birds from any given 

 locality are constant, and the lowland form was differen- 

 tiated under the name of Gallh-allus hectori reischeki (Iredale, 

 Austral Avian Record, vol. ii. 1913, p. 15). 



