Movements of the Ground. 191 



The levels rest on a mass of masonrj, built twenty years 

 before the observations were begun, but leaning against the 

 outer loest wall of the house ; they are at a depth of more than 

 a metre below the ground outside, and at a distance of a 

 metre from one another. M. Plantamour does not think that 

 the movements indicated by the levels are due to oscillations 

 of the outer wall under the action of changes of temperature. 

 But we cannot be certain that they are not partly so caused, 

 and it is impossible not to regret that observations under- 

 taken with such care and perseverance should not have 

 been made on levels completely isolated from all external 

 accidents. 



One of the levels is orientated east and west, the other 

 north and south. During the first two years, however, owing 

 to an error in the compass employed, their directions were 

 7° S. of E. to 7° N. of W., and 7= E. of N. to 7° W. of S., 

 respectively. Readings were taken five times a day (at 

 9 A.M., noon, 3, 6, and 9 p.m.) during the first year, in order 

 to determine the nature of the diurnal oscillations ; and 

 afterwards twice a day, at 9 A.M. and 6 P.M., that is, a little 

 after the daily minimum and before the daily maximum. 

 The mean of these readings is taken as the value of the in- 

 clination for the day. 



Each year of observations begins on October 1, but, during 

 the first year, the N.S. level was not ready until December 23, 

 1878. Tables are given in each memoir of the mean inclina- 

 tion of both levels for every day in the year, and the results 

 are graphically shown on a plate, as well as the curve of the 

 mean temperatures recorded at the observatory of Greneva. 



Results of M. Plantamour' s Observations. — 1. In the E.W. 

 level, a rise or fall of the external temperature produces a 

 rise or fall, respectively, of the east end of the level : the 

 movements of the ground lagging behind the changes of 

 temperature. The extreme oscillations are determined by 

 the continuity of a certain mean temperature rather than by 

 the actual maxima or minima. The conclusion is therefore 

 natural that variations of external temperature may be one 

 of the principal causes of the periodic movements of the 

 ground. 



2. The period of oscillation experienced by the N.S. level 

 is also annual, but its range is much smaller. In a general 

 way, the south end follows the changes of external tempera- 

 ture, that is, it is lowered in winter and raised in summer. 

 But the temperature variations of shorter period produce an 

 inverse effect ; as the temperature rises the south end is 



