1920 .] 
Departmental Report. 
173 
found to have incised deep gorges in the ancient valley-floors, but the smaller 
streams—for example, Sloven’s Creek—have cut gorges only near their 
junctions with the larger streams, and higher up flow in shallow channels 
cut in the gravels that cover the hard rock floors of the valleys. The main 
valleys, and in many places the subsidiary valleys also, are bordered by 
great gravel-veneered terraces, some of which are 600 ft. or more above the 
present streams. Although the valleys have been partly formed by stream 
erosion and to a smaller extent by ice erosion, block-faulting accompanied 
by warping has had a great influence in the evolution of the present 
topographical forms. 
The prevailing rock-formation throughout western Canterbury is a great 
series of greywackes and argillites, generally supposed to be of Trias-Jura 
age. In the foothills district there is a considerable development of coal¬ 
bearing Cretaceous and Tertiary rocks, which lie upon the greywackes with 
most uncompromising unconformity, and in places are associated with 
various types of igneous rock in the form of dykes, sills, and flows. West of 
the foothills region are isolated patches of Cretaceous or Tertiary rocks, 
most of which have been mapped bv von Haast.* These patches owe 
their existence to down-faulting and down-warping, which have saved them 
from erosion. The only other rocks of the mountain region are Pleistocene 
and Recent gravels, the former in part of glacial origin. 
The coalfield which is about to be worked by Mount Torlesse Collieries 
lies in the valley of Broken River, one and a half to three miles south-west 
of Avoca Railway-station. To the south is Mount Torlesse (6,422 ft.), to the 
north the isolated mountain No Man’s Land (3,662 ft.). The coal-bearing 
area appears to be about two square miles, half of which is on the north sid.e 
of Broken River between Rocky and Winding streams, and half on the south 
side. The coal-measures consist of a basal conglomerate, usually thin and 
in places absent, resting on greywacke, and followed by a considerable thick¬ 
ness of light-coloured sandstone—several hundred feet—which in its lower 
part is interstratified with bands of carbonaceous shale and coal. The sand¬ 
stone gradually passes into greensand, of which there is also a great thick¬ 
ness. Near the top of the sandstone there is in places a band or bands of 
more or less shelly conglomerate. The shells found here and in a similar 
conglomerate north of No Man’s Land include Inoceramus sp., Ostrea sp.= 
“ black oyster ” of Hector and McKay, and casts of Conchothyra parasitica 
(McCoy MS.) Hutton. These fossils clearly indicate that the coal-measures 
are of Cretaceous age. In the Castle Hill district (Trelissick Basin) similar 
beds, also with Cretaceous fossils, are succeeded by marls, limestones, 
tuffs, sandstones, &c., which from the lowest tuff band upwards contain a 
Tertiary fauna. Speightf states that layers of volcanic fragmentary 
matter, associated with beds of calcareous material, occur near the junction 
of Sloven’s Creek with Broken River. A basic dyke (mentioned by Speight) 
crosses Iron Creek rather less than a mile above its junction with Broken 
River. The dyke is perhaps 25 ft. thick, and appears to strike westward 
on one side of the creek and north-eastward on the other. A similar dyke 
is seen intruding a large coal-seam on the south side of Broken River a 
short distance above the mouth of Iron Creek. The alteration produced in 
the coal will be described later. Both dykes consist of dark fine-grained 
* See map accompanying Geology of Canterbury and Westland, 1879. 
t R. Speight, The Stratigraphy of the Tertiary Beds of the Trelissick or Castle 
Hill Basin, Trans. N.Z. Inst., vol. 49, p. 343, 1917. 
