1 42 Proceedings of the Royal Society of Victoria. 



of a piece of wood of uneven texture, now working fairly 

 smoothly, now catching in a harder band, wrenching a piece out, 

 and again proceeding evenly. An indication of the direction of 

 the ice can be thus obtained. In this case the ice appears to 

 have come from ten degrees south of east, but it would be unwise 

 to infer much from this one instance, especially as there is 

 abundant evidence to show that the Silurian rocks have under- 

 gone considerable movement since this ancient ice-age. 



Wherever we have seen the junction of the till with the 

 Silurian, the former has always been intensely hard and unstra- 

 tified, and the latter invariably grooved, smoothed, and sti'iated. 

 These are facts, which, in our opinion, point to one conclusion, viz., 

 that the lowest member, at least, of the glacial series is morainic, 

 due to the action of land ice, of the former presence of which 

 we have unquestionable evidence in the roches jnoutonnees. It will 

 have been seen that the Bacchus Marsh sandstones must be 

 considered as part of the glacial series — a conclusion to which our 

 friend, Mr. Brittlebank, has also come independently of us. As 

 the only fossils obtained so far are plant remains, a fresh-water 

 origin for them is indicated, and it is reasonable to suppose 

 that these sandstones were deposited in a glacial lake in which 

 floating ice drifted. The clay bands in the sandstone may perhaps 

 have been formed by subglacial material cai'ried into this lake by 

 streams. Any floating ice would be drifted with the currents and 

 drop their burdens occasionally in the accumulating silt. Such a 

 lake may have been almost an inland sea. The vast size attained 

 by glacial lakes in America during the last ice-age is well known. 

 The alternation of boulder beds with plant-bearing sandstones is 

 only what would be expected on the astronomical theory of ice- 

 ages. 



POSTSCRIPT. 



Since reading the above paper we have discovered several 

 beautiful examples of roches motifonnees, near Coiraadai. The 

 smoothed and grooved surfaces can be traced right beneath hard 

 unstratiiied till. There is also good evidence to show the direction 

 the ice took at this localit^'^, viz., from S.S.W. to N.N.E. These 

 are by far the best example of roches nioutonnees we have seen in 



