McLeod. — Some Caves and Water-passages. 479 



canic rocks of a greater age are known to occur iii other 

 parts of Southland at no great distance, it is not unHkely 

 that they underlie the rocks of the Hokonui system at Nugget 

 Point, and the presence of derived fragments in a dyke pene- 

 trating them would not be at all improbable. 



The date of the intrusion of the dyke can be fixed with 

 tolerable certainty. It is found in Triassic rocks, so that it 

 cannot be earlier than that period. Further, I found near it 

 a derived sedimentary rock which contained a small Monotis. 

 If the rock was really formed from the dyke it fixes the date 

 for the close of the Triassic period, or not later than early 

 Jurassic. There is some difficulty, however, in determining 

 accurately the origin of the derived sandstone, as the sand- 

 stones in the immediate neighbourhood are highly feldspathic, 

 but contain a large proportion of plagioclase. However, I did 

 not notice this in the derived volcanic sedimentary rock. A 

 granite intrusion occurs in the neighbourhood at Omaru 

 Creek (see paper by Captain Hutton, read before the Royal 

 Society of New South Wales, 1889). This is on the same 

 line of country, and from the description of the rock there 

 are marked affinities to the one at Nugget Point. They may 

 therefore be of the same age. 



Abt. XLI. — Some Caves and Water-passages in the Grey- 

 mouth District. 



By H. N. McLeod. 

 \^Read before the Wellington Philosophical Society, 1st July, 1903.] 



In the following note will be found illustrations of the 

 action of water in forming caves and underground passages in 

 two different ways, in the one by percolation from a ridge-top, 

 and in the other by overflow from flood-waters. 



The latter instance is to be found in the vicinity of 

 Kumara, on the left bank of a stream known as Whisky Creek, 

 which obtains its name from its flowing by the site of a 

 raided still. Here, a few chains above the point where the 

 track crosses the bed, the flood-waters pass into an under- 

 ground passage, which, after expanding into a vault, branches 

 into other passages. The outlets to these were not traced, but 

 without doubt drain into the Eiver Teremakau, which is not 

 far distant. An entrance was forced through debris which 

 almost closed the entrance, and a distance of a hundred paces 

 was traversed, when further progress was barred by the 

 silting-up of this passage. Return was made to the vault, 

 and another passage traversed for fifty paces. The descent of 

 this was not without some difficulty, owing to sudden drops 



