110 



mainland. That its extension is in a westerly direction 

 rather than southerly, parallel to the Mount Lofty Ranges, 

 cannot be due primarily to the original Palaeozoic folding, 

 the axis of which also bends in a similar fashion ; but it may 

 be due to it, secondarily, in that the bounding fault-scarps 

 have developed parallel to the folding planes of the rocks, 

 as in the case of the Mount Lofty Range itself. The most 

 obvious explanation of the passage is that it is a senkungs- 

 feld, i.e., an area dropped down between two fault-planes, 

 respectively the southern scarp of the main range and the 

 northern of the island. The Pages might be considered as 

 the tops of a sunken residual. But in the case of the 

 Inman, Hindmarsh, and Upper Finniss Valleys it is clear 

 that their great maturity is due to the fact that they were 

 carved by the Permo-Carboniferous glaciers and filled with 

 their soft till. This has been quickly removed when first 

 exposed to the attack of streams, rejuvenated by the uplift. 

 Might it not also be suggested that the Backstairs Passage 

 was a wide glacial valley filled with till, which has been 

 subsequently almost entirely removed by stream and marine 

 erosion ? Several facts are in support of this. The base of 

 the valley must, of course, have been below sea-level ; but 

 so is that of the Inman glacier at Victor Harbour. The 

 researches of Mr. Howchin < 10 ) have shown the strongly 

 glaciated nature of portion of the southern scarp of the main- 

 land, and he has proved the presence of glacial boulders near 

 Cape Jervis/ 11 ) He has also described Permo-Carboniferous 

 glacial till on northern Kangaroo Island. < 12 ) 



The depression is thus bounded on both sides by glacial 

 material and, in places, striated surfaces — facts strongly in 

 support of the second hypothesis. It is, of course, possible 

 that block- faulting may have assisted in the formation of the 

 passage, but the author's inclination is to give it a minor 

 role. On the glacial hypothesis The Pages should be roches 

 moutonnees. The Admiralty soundings are of little help in 

 deciding the question, as they show only that a flat bottom 

 exists in the passage at a depth of less than 20 fathoms. 



The drainage alterations during the various periods of 

 earth movement require much further study. Rivers were 

 captured, as the heads of the Onkaparinga by the Torrens, 113 ^ 



<10) Trans. Roy. Soc, S.A., vol. xxxiv., 1910, p. 1, pis. i. to 

 xvii. 



(ii) Rep. Aus. Asso. Adv. Science, vol. vii., 1898, p. 124. 



(12) Trans. Roy. Soc, S.A., vol. xxiii., 1899, p. 198, pis. iv. 

 and v. 



(13) This conclusion, though reached independently by the 

 writer, has been, he finds, Mr. Howchin's view for some time. 



