BY L. A. COTTON. 807 



indicates that the flow was probably in that direction. Amongst 

 the quartz-pebbles in the wash are to be found a number of 

 blue-coloured rocks, subangularor ellipsoidal in form, and ranging 

 from 2 to 24 cm. in length. This particular band of rock is 

 20 cm. in thickness, and lies between gravel and drift of normal 

 character. The existence of these bands of wash, at different 

 levels in the drift, is doubtless due to the action of minor floods. 

 Andrews has shown that the work of aggradation is performed 

 in tlie interflood period. Each moderate flood excavates some of 

 the material aggraded during the interflood period. The largest, 

 or true flood, removes the whole of this material, and corrades 

 the channel. As the flood suV)sides, aggradation commences; 

 matter held in suspension gradually settles; the coarser material, 

 on account of its greater weight per unit of surface, sinks first, 

 and thus a grade from coarse to fine material is established. 

 Pebbles of greater density associate with larger stones of less 

 density; and, finally, sand covers the whole. The effect of a 

 moderate flood, on the deposit, following a large flood, will now 

 be considered. The moderate flood has not sufficient energy to 

 remove the whole of the material deposited after the large flood, 

 but it can remove some of it. As this moderate flood begins to 

 subside, deposition takes place, and another series is formed 

 having the coarser pebbles at its base. Thus it is possible for 

 many zones to be formed. The coarser pebble-zones formed in 

 this way constitute the layers of wash. The lowest layer of wash 

 is usually the most payable, as the diamonds are larger and more 

 abundant. The layers of wash are somewhat irregular. The 

 bottom-layer follows the contours of the channel-base, which is 

 generally very uneven. The upper layers of wash, on the other 

 hand, conform very uniformly to the horizontal, as they are built 

 up on the loose sands and gravels, which readily adjust them- 

 selves to the level under the influence of running water. Yet 

 there is a degree of irregularity : the wash seems to be cut off 

 very abruptly in places, and no clue seems to be aflforded by 

 which it may be traced. A very striking illustration, which 

 may explain this irregularity, was observed in the bed of Cope's 

 Creek, near its junction with the Gwydir. A recent flood of 



