38 THE PROBLEM OF ANTIQUITY OF MAN IN AUSTRALIA 
Ballan and the Bacchus Marsh basin, a distance of eight miles as 
the crow flies or eleven along the course of the river, its bed falls 
95 ft. per mile’ and the river flows through a gorge ranging from 
400 ft. to 600 ft. in depth. Possibly in its early development, when 
the stream was cutting its valley into the Newer Volcanic lava, it 
flowed over a waterfall into the Bacchus Marsh basin, and owing to 
erosion the ledge over which it fell retreated upstream and finally 
disappeared, leaving a steeply inclined river bed. Rainfall may 
formerly have been greater than it is now. 
For a period of five years Brittlebank (1900) conducted experi- 
ments at numerous points to determine the rate of erosion of the 
bed of the gorge, and his results work out at 0°58 inch per century. 
The period of experiment was probably too short and his methods 
not sufficiently accurate to give a figure other than an approxi- 
mation to the right order of magnitude. No data are available 
for similar streams elsewhere. The average rate of degradation 
of the Mississippi basin is estimated at 0°34 inch per century, and 
Niagara Falls are retreating upstream at an average rate of 
4 ft. 6 in. per annum (Chamberlin and Salisbury, 1905), but the 
general fall per mile of the Mississippi is low and the rate of 
retreat of Niagara Falls is exceptionally rapid, so these figures 
are of little value in an enquiry into the rate of formation of the 
Werribee gorge, but they indicate that a waterfall causes rapid 
erosion. 
If we assume that the rate of erosion of Werribee gorge during 
its whole development was very high and averaged 20 times the 
amount that Brittlebank’s figure indicates for the present time, 
say 12 inches per century, the excavation of the gorge would take 
60,000 years. Though this figure is hypothetical and probably 
too small, it indicates that the sub-basaltie gravel bed at Myrniong 
in which the implements were found is Pleistocene in age. 
In the Great Buninyong Estate mine, near Ballarat, Victoria, 
fragments of bones of extinct marsupials were found 240 ft. from 
the surface in black pyritic clay underlying a basaltic lava flow; 
the bones were near the base of the lava. The clay is a swamp 
deposit and the basalt is a lava flow from Mount Buninyong, a 
scoria cone in the vicinity (Hart, 1899). The bones, which are 
mineralized and impregnated with pyrites, were identified by 
De Vis (1899) as those of Diprotodon and an extinct kangaroo, 
Macropus faunus. One fragment, probably part of a Diprotodon 
rib, is about 6 inches long, irregular at one end and terminated 
at the other by two cuts from opposite sides which do not meet 
10. Figures for drainage area, discharge and fall of river bed were supplied by the Victorian 
State Rivers and Water Supply Commission: river gaugings at Bacchus Marsh were taken 
over a period of 15 years. 
