368 Proceedings of the Roijul Socletij of Victoria. 



we need seek no explanation of how a basin of this depth could 

 have been formed and filled. Local opinion frequently refers it 

 to volcanic explosions, of which, however, there are no indica- 

 tions at all. There are no volcanic fragmental rocks, and the 

 lava flows do not originate here. 



On the other hand, a number of observations favour the view 

 that there has been here a local subsidence. In the valley of 

 Sailor's Creek, the next valley, at a point north-west of the basin, 

 there is marked on the geological map a broad " dyke " con- 

 nected by dotted lines with the large Cbrinella dyke at 

 EganstoAvn. The gravel in the creek bed conceals much of this 

 area, but wherever I have seen it, it consists of a greyish paste 

 in which are irregularly distributed fragments of the Ordovician 

 bedrock, often of considerable size. It is a belt of fault rock, 

 of considerable width, perhaps accompanied by some intrusive 

 material, though I have seen none here. An undecomposed 

 basaltic dyke occurs a short distance to the south. This belt 

 of broken rock would naturally not be detected on the hill- 

 sides, but conversations with the miners give some support to 

 the view that it extends towards the basin and occurred in the 

 tunnel workings between the place where it is seen and the 

 basin. Similar broken ground seems to occur also in workings 

 in a gully west of Sailor's Creek. At Eganstown the Corinella 

 dyke is shown on the map with a width of 10 or 12 chains. 

 At its western end it is surmounted bv scoriaceous material, 

 forming a low hill ; further west in the Deep Creek it appears 

 only as a few thin dykes. East of Eganstown it is exposed 

 with much diminished width in a quarry. A few decomposed 

 dykes are seen in the road cuttings west of Sailor's Creek, and 

 some distance further ea.st beyond the basin similar dykes occur 

 in the cuttings of the Ballarat railway and in the creek bed 

 east of the Jubilee Lake. Continuing, the line leads to 

 Wheeler's Hill, an old point of eruption. If the line were con- 

 tinued to the west it would pass through Mount Moorookyle 

 and McDonald's Hill and others beyond, but the number of 

 these volcanic hills makes coincidence less important. The 

 basin appears to be situated on a line of weakness at the time 

 of the volcanic activity. It lies directly on a line between the 

 hill on the dyke at Eganstown and "Wheeler's Hill, and about 



