226 J. E. Macnellan, C.E.— On [Oct. 26, 



rammed till all the interstices between the culvert and the excavation 

 above the springing of the arch were filled in. After the pipes and 

 valves were properly fixed, the portion of the culvert from the tower 

 for a distance of To feet was built in solid, the material used being 

 piincipally concrete with a little masonry. This plug was made 

 longer than is usual, on account of the shaly nature of the ground 

 below the culvert. The depth of this shale, as I before mentioned is 

 unknown, a .50-feet bore having failed to pierce it. It is not unlikely 

 that the water from the reservoir will soak through this shale and 

 issue out at some point or points beyond the bank. 



Advantage has been taken of a nek some distance from the bank to 

 construct the overflow. The ground was excavated and levelled for a 

 width of 150 feet, which is calculated to be amply sufficient to carry 

 off the surplus w^ater. 



When the water in the dam is on a level with the bottom of the 

 20-inch pipe, that is, is empty, the area of what remains in the kolk is 

 34 acres. This can only be used for watering stock. The extent of 

 land covered by water when the dam is 4 feet deep at the tower, is 

 500 acres, and at present (March 1886), with a depth at tower of 6\ 

 feet, it is at least 800 acres.* 



Beacons have been erected at suitable points outside the dam so 

 that its area can be easily and expeditiously measured at different 

 levels. This will subsequently be of great value, as a close approxi- 

 mation can be made as to the amount of water available for irrigation 

 and other purposes. 



Careful and regular measurements have been taken of the depths 

 of the water in the dam to find out, as far as possible, the rate of 

 evaporation and soakage. 



No water at all was taken from the dam in 1884, so the amount as 

 shown in the annexed table represents exactly what w^as lost by 

 evaporation and soakage during that period. 



Irrigation was first commenced in July 1885, and has been con- 

 tinued up to the present time. 



The amount of land under cultivation is about 25 acres, but the 

 water used for this patch bears a very small proportion to that lost by 

 evaporation. In a whole day's leading, no appreciable difference in 

 level of reservoir could be detected, the loss during the year 1885 may 

 be considered as almost entirely from evaporation and percolation. 

 The observations of the two years can thus be compared, from which 



* No chance to measure yet. 



