Haast-—On an dneient Manufactory of Stone Implements. 151 
At that time it doubtless formed part of the mainland, as shown by the 
kitchen-middens and rude stone implements, flakes, and cores appearing 
exposed on both sides of the nearly vertical cliffs, in positions corresponding 
with each other. 
This locality is the more interesting, as it belongs doubtless to an inter- 
mediate period, when the Moa had already become extinct, and when 
possibly cannibalism had begun to be first indulged in. 
The Otokai kitchen-middens are, therefore, different from those existing 
near the mouth of the Kaikorai Creek, some six miles to the north, situated 
amongst the remarkable sand-dunes, which cover here an area of more than 
a square mile, and ascending to an altitude of 300 feet on the southern 
slopes of Otago Peninsula. 
At the.foot of these sand-dunes, and fronting the northern banks of the 
Kaikorai Creek estuary, a well-defined line of kitchen-middens lies about 
five feet above high-water mark, having a thickness of from several inches 
to more than one foot. 
These kitchen-middens consist mostly of shells, of which Chione stutch- 
buryi and Mesodesma nova-zealandie are the most numerous. It is remark- 
able that these shells are nearly twice the size of those now inhabiting the 
Kaikorai estuary. In addition, Mytilus smaragdinus is well represented, but 
it is rather smaller than the same species found at present near the coast 
close by. Amphibola avellana and some others are also occurring in more 
or less large quantities. 
Amongst these shells Moa bones are scattered here and there. 
They are broken, often burnt, and have doubtless been deposited con- 
temporaneously with the shells. It is thus evident that the Moas had 
already become so scarce that they only occasionally could be obtained, and 
the natives had to look towards getting other food as a regular means of - 
Jm. However, it is to be expected that more towards the centre of 
e sand-dunes, older deposits proving human occupancy exist, and which, 
as in other localities, will consist almost exclusively of the remnants of the 
extinct Dinornithide. Stone implements in the same locality are not scarce. 
They consist of very rude adzes and knives, mostly chipped from basaltic 
boulders obtained in the neighbourhood; however, similar tools made of 
flint, chert, quartz, and chalcedony are also represented. 
Some few perfect and more numerous broken polished stone implements, 
together with whetstones and other polishing materials, were also obtained. 
Although I could devote only one day to an examination of this interesting 
locality, I was enabled to obtain a good insight into the character and 
position of the kitchen-middens under review, being fortunately guided by 
Mr. F. L. Jeffcoat of Winchendon, who lives close by, under Stony Hill, 
