In an earlier communication to this Society it was shown that there is a slight excess of frequency of earthquakes, above the general average, during the day in summer and during the night in winter. It was also pointed out that this might possibly be a result of the unequal range of variation of the gravitational stresses set up by the sun, and, if so, might be expressed in the form that there was an excess of frequency of earthquakes during that half of the day which contained the upper meridian passage when the declination of the stress-producing body and the latitude of the place were of the same name, and during the half which contained the lower meridian passage when they were different. If this were the true statement of the variation, a similar effect ought to be noticeable, and of greater amplitude, when the record was tabulated by lunar time and declination. Such was actually found to be the case in the only record already tabulated in a form permitting the application of the test, but the number of shocks contained in this record was not large enough to give more than very small weight to the numerical values obtained from it. Since then I have been able, with the assistance of the Government grant of the Royal Society, to prepare a fresh tabulation of the records of Italian earthquakes by lunar, as well as by solar, hour-angle and declination. The period chosen was the 19-year lunar cycle