It is well known that during the last few decades since the classical case history of reservoir induced seismicity at Lake Mead, U.S.A., there bave been about two dozen cases uptil now (Rothe’, 1970; Proc. COSERI, 1973; Guba ot al, 1974), wherein significant seismic activities followed reservoir impounding. Though, in most cases, largest earthquakes observed following reservoir impounding were below magnitude 4.5, in few cases such as at Koyna. Ka.riba, Kremasta and Lake Mead, the magnitudes of largest earthquakes recorded were higher, even upto 6.5 (Koyna .earthquake of 10.12.1961). It is evident from the above case histories that the number of reservoirs exhibiting seislDlc activities following impounding is insignificant compared to their total number as such. It is quite pJssible that seismic activities at much lower level could be more prevalent than observed so far and had escaped detection due to lack of close and proper instrumental observations near reservoirs. In a projlramme initiated immediately after the Koyna earhquake of December 10, 1967 to investIgate existence of such low level seismic activities at various virgin reservoir sites in the Deccan trap region, seismographs suitable to record ultra-microearthqukes down to magnitude -2.0 were installed near some newly impounded reservoirs. Though a few of the reservoirs did not show even ultra-microearthquake activity, Mula reservoir situated in Deccan Trap region exhibited u1tra-microearthquake activity significantly and the same did ftuctuate with water level variations in the virgin reservoir.