278 T. S. Hart: 



Hank of tlie higher ground not far from suveral places where the 

 east fall is steeper than usuaL 



Beaumaris J3ay is no (k)ul»t due to marine attack favoured by 

 tlio struetuial eliaraetei'. Furtiier to the south-east the sea has 

 liad to give phiee to the material brought down from the east 

 side of the Cheltenham axis, and by the Dandenong Creek. Recent 

 marine fossils oei-iir at the moutli of the main <lrain at the head 

 of tile creek at Mordialloe just above sea level. 



The relation of the rocks to the valleys. 



The tertiary rocks may be divided for tlie present purpose into two 

 parts — the upper, very weak sandstones, usually pale coloured, '^ 

 but occasionally more ferruginous and then stronger ; the lower 

 usually brown and often more clayey, with ironstone bands and 

 some strong coarse ferruginous beds. The upper parts are highly 

 permeable, they readily lose their iron cement in weathering; pro- 

 bably even the more ferruginous and stronger beds yield fairly 

 easil3' to soil waters. These form the bulk of the higher cliffs, and 

 are exposed in every railway cutting of any depth in the tertiary 

 areas. The lower less permeable beds are seldom touched in the 

 artificial cuttings, l)ut probably form the floor of many of the 

 valleys. They are seen at many points on thi' coast, and some of 

 the springs can be seen to occur at their upper edge. 



On a catchment consisting of the upper parts of the tertiaries, 

 a very large i)roi3ortion of the rainfall must be absorbed, and if 

 the valleys have not reached less permeable beds, may never reappear 

 on the land surface in the vicinity. The erosive power of the 

 stream woidd be con-esjtoudingly reduced, and when, in addition, 

 tliere are small catchments, low grades, and only a moderate rain- 

 fall, the streams nmst be Vi'vy wrak. When the stream reaches the 

 less permeable lower beds it will rei-eive additions from springs. 

 These springs must produri' a local weakening, and hence greatly 

 iiicrease th(> j)owei' t(i widen thi; valley oi' to cul out a channel of a 

 oraiich gully nu llu- line of a spring. A steeper slopi- caji be seen 

 sometimes behind a spring at the foot of a hill. This 

 power of widening has already been referred to, and is 

 no doubt of general application. It will evidently give 

 increased it(,)Wer to tlie stream whiili liist reaches the 

 less ))erineable beds, that is, other things l)eing equal, 

 to the stream on or near the anticline. A somewhat similar effect 

 could be produced by any stream reaching the permanent water 

 level, but in this district the two causes are i)ri)bably practically 

 coincident. 



