Page:The Kea, a New Zealand problem (1909).pdf/147

Rh The land thus seems to have been elevated and again depressed, leaving it very much in its present condition.

This closer connection between the two groups of islands may explain the presence of Kea fossils on the Chathams.

This theory, however, only adds mystery to the strange fact that no Keas or Kea-fossils have ever been found in the North Island, situated only 15 miles away.

As early as 1888, Sir W. Buller says that he is certain that these interesting birds would soon be extinct, but in spite of the thousands that have been killed they are still common in the mountainous country of the South Island. No doubt the almost inaccessible position of their nests, and the rough nature of the country in which they live, are responsible for their preservation.

When harassed they often retreat to the most inaccessible fastnesses of the Alps. Here they are practically safe, for this exceedingly rough country can never be of much use except for scenic purposes. It is, therefore, doubtful if the Kea will become extinct for many years to come.

If, however, closer settlement of the land, accompanied by the destruction of the forest and the systematic slaughter now going on, should threaten to exterminate the Keas, I would suggest that, in order to prevent these interesting birds from becoming absolutely lost to the scientific world, a number of them should be placed on one of the outlying islands, where they could live and flourish without doing injury to any one. The most suitable islands, as far as I can ascertain, are the Aucklands, which lie 190 miles south-by-west from the most southerly point of Stewart Island. There would be very little chance of the birds returning to the mainland; and though the hills rise to a height of about 2,000ft. only, there seems to be enough forest and high country to make a very satisfactory reserve for these interesting parrots.

Though the sheep are favourite objects of the Kea’s attack, it does not seem to confine itself to them, for several instances