328 Transactions. — Geologij. 



Towards the close of the loess record two new characters 

 are introduced, the mterpretation of which for a long time 

 puzzled me completely. From 2ft. to 4ft. beneath the sum- 

 mits of the ridges the loess shows in weathered road- and rail- 

 w^ay-cuttings a projecting baud 1ft. to 3ft. thick, the material 

 in which differs from that above and beneath it only in being 

 more compact, and thus resisting the weather better. The 

 presence of drought-cracks nearly everywhere has enabled the 

 rain to carve this band into elongated bosses. On the slopes 

 of the spurs this character gradually gives place to another, 

 which is quite unique in the formation. Where well developed, 

 as on the slopes of the larger gullies, this consists of hard flaky 

 layers, some rusty, some not so, some even whiter than the 

 loess generally ; the whole generally but a few inches, but in 

 some places a few feet, in thickness, and frequently, or rather 

 generally, separated into small roughly cubical fragments. The 

 whiter portions look just like the "pugged " clay formed be- 

 neath landslips and seen where such slips have been sectioned 

 in roadwork. The two related characters show more or less 

 clearly in every spur in cliff and cutting at the coast, and the 

 pugged layer in most of the sidling road-cuttings on the slopes 

 of the gullies all over the plateau. The only explanation I 

 can find to account for these characters is that they register 

 the phase of greatest severity of the ice age to which the loess 

 belongs — the second glacial period ; that they show that the 

 summits of the ridges at that time were compacted by a heavy 

 load of ice, while the surfaces of the slopes were "pugged" 

 by the ice creeping over them to form ice-streams in the gullies. 

 Adopting this explanation, the extensive denudation of the 

 dolerite and underlying gravels seen in the larger gullies and 

 their branches becomes comprehensible ; whilst it must be 

 simply a cause of utter bewilderment at the tune required, if 

 we must believe this denudation was effected by the trifling 

 surface-drainage now at work. 



Knowing that in the discussions which took place some 

 years ago upon the glaciation of New Zealand the view was 

 strongly cornbatted that there had ever occurred such a degree 

 of glaciation as to involve the lowlands of Canterbury, I have 

 been the more cautious in adopting it. It is, however, well 

 supported by plainer evidence in the hnmediate neighbourhood. 



The map and descriptions of the icefields and glaciers of 

 the great glacier period, in Haast's " Geology of Canterbury 

 and Westland," can by no means show the full extent of the 

 glaciation of Canterbury. The map shows, the text describes, 

 only the larger glaciers originating in the Southern Alps. 

 ^Besides these there must have been many minor snow-fields 

 with their glaciers. One of these minor fields evidently 

 existed on the north side of Mount Misery (otherwise Cave; 



