198 



within the last few years." These two lakes are to all 

 appearances much shallower than the other lakes. 



The graph shown in fig. 9 has been carefully compared 

 with the corresponding rainfall records of the south-eastern 

 counties of South Australia. ♦ This rainfall is approximately 

 the same as that of the chief intake beds of the basin. There 

 is no close correspondence to be noted at first sight between 

 the average annual rainfall and the variation of water level, 

 but general relations are revealed by careful examination. 

 The two wet years of 1909-1910 were, for instance, followed by 

 a corresponding considerable rise in water level. The con- 

 sistently low water level of the years 1902-1910 corresponds 

 with the three dry years of 1902, 1904, and 1907. The 

 great drought of 1914 was closely followed by a distinct and 

 rapid lowering of the lake levels. The attempt to correlate 

 the rainfall and variation of lake levels is, however, far from 

 satisfying. 



Apart from these major movements, as revealed in the 

 graph, the detailed weekly records of water level show 

 a minor annual movement that is of much interest. This 

 consists in a general rise of level in the summer (December- 

 January) months of the year, and a general lowering of level 

 in the winter (June- July) months. This is obviously not due 

 either to extra local consumption of water nor to increased 

 evaporation/ 21 ) since the rise in level corresponds with the 

 increase of both these factors. The area is one of winter 

 rains, and the obvious conclusion is that the summer rise is 

 due to the arrival along underground courses of a previous 

 winter increment of rain received in the main intake beds. 

 Whether it is the previous winter's rains (six months before), 

 or those of a season prior to that, cannot be discovered with- 

 out more detailed figures and further investigation. It seems 

 clear, however, that the rise or fall of the lake levels depends 

 on the amount of rainfall either six or eighteen months prior 

 to such rise, and argues either of those periods as the time 

 taken for the water to travel underground from the main 

 intake areas to Mount Gambier. 



Further Note on the Correlation of the Rainfall with 

 the Variation in Level of the Lakes. — Since writing the fore- 

 going note, regarding the relationship between the rainfall 

 of the south-eastern counties of South Australia and the 

 variation in level of the Mount Gambier lakes, the rainfall 

 records of the County of Lowan (Vict.) have been obtained 



(21) The average evaporation per annum at Mount Gambier 

 is 35 inches (vide Special Report on Lakes Leake and Edward, 

 Ann. Rep. Govt. Geologist, S. Austr., for 1917, p. 14). 



