season, or during the driest year, would not necessarily give the 



lowest yield; for if the season or year of drought followed a period 

 of heavy rainfall, the water table, having been raised during the 

 rainy years, would still deliver during the period of drought, • 

 large amount of water that had fallen during the previous years. 

 Consequently the residual mass curves were worked up. The 

 general agreement between these mass curves and the fluctuations 

 in the elevations of the ground water justify their computations." 



The importance of the Lake George record was recognised 

 by Mr. Russell very early in his investigations, and in his 

 paper of 1876 lie supplied information concerning its fluctu- 

 ations from the date when the lake was discovered in 1820 

 by Sir Thomas Mitchell. On 1st December, 1886, he read a 

 paper before this Society entitled "Notes upon floods in 

 Lake George," in which he described all the work done by 

 him in his endeavour to ascertain the level of the water in 

 the lake from time to time. The records of the Lands 

 Department were searched to enable the margin of the 

 water to be determined from old surveys. Further inform- 

 ation was obtained on the spot by himself from old residents, 

 and the diaries of explorers and settlers were also looked 

 into, in order to obtain information. Finally a surveyor 

 was instructed to connect all the points obtained by direct 

 levelling. In this manner Mr. Russell was able to con- 

 struct a diagram, which he appended to his paper, showing 

 the conditions of the lake with approximate accuracy from 

 the year 1820 to 1885, in which latter year lie established 

 a self registering instrument similar to an automatic tide 

 gauge, such as we have at Fort Denison, which has regis- 

 tered accurately the fluctuations of the water up to 1907, 

 since when they have been unfortunately discontinued. It 

 is to be hoped, however, that the authorities will recognise 

 the very great importance of continuing the record. The 

 small expense of keeping the register, and attending to the 





