30 



TUTIRA 



towards the east. It was coated with a matted humus overlying 

 seolian pumice grit, overlying disintegrated red sand. Filtering 

 through the loose surface mould on to the layer of grit and sand, 

 rain - water did not so much penetrate downwards as follow sub- 

 cutaneously the trend of the slopes. It passed away in an unseen 

 filtration ; I imagine it lipping from grain to grain of grit, moving as 

 evenly, as diffusedly, and with as little current, as water spilled on an 

 inclined sheet of blotting-paper. Finally, at the base of our imaginary 

 slope the accumulated soakage would gather against the cliff of the 

 next " comb," against, that is, the next western-facing precipice the 

 next segment of canted plateau. There, unable to press further east- 

 ward, it would be temporarily held. Eventually, at the base of our 



Imaginary section of Central Tutira prior to erosion, the dotted lines showing yet 



hidden interstices. 



block, there would develop as an essential factor in our scheme of 

 erosion, a tiny runnel or rivulet escaping north or south towards one 

 or other of the main rivers, towards the Waikoau, the Matahorua, or 

 the Waikari. Up to this point our imaginary section of hill-slope has 

 seemed a homogeneous whole, but now, with the deepening of the 

 rivulet a deepening sympathetic with that of the whole drainage 

 system of the station two different qualities of surface would begin 

 to reveal themselves. The little stream would ooze at the earliest date 

 along the humus ; as the deepening process continued, it would trickle 

 alongside the layer of volcanic grit, and later along beds of disintegrated 

 sands. With a prolongation of the process our rivulet, cutting still 

 deeper, would at last begin to skirt alternately two different qualities 

 of material, hard and soft the " teeth " and " interstices " respectively of 



