192 Mr. C. Davison on the Periodic 



lowered and vice versa. This anomaly is never observed in 

 the B.W. level, and its cause is unknown. 



3. Diurnal variations are exhibited by both levels. In the 

 B.W. level they are frequent ; they vary much in intensity, 

 being sometimes insensible, but at other times very pro- 

 nounced, amounting to as much as 3"-2 (on Sept. 5, 1879). 

 In the N.S. level they are rare and always small. When 

 they occur in one level, moreover, they are absent in the 

 other*. In the E.W. level M. B. Plantamour, applying 

 Bessel's formula for periodic phenomena, finds that the 

 maxima and minima of elevation of the east end generally 

 fall between 6 and 7| p.m. and a.m. 



4. Besides the periodic movements indicated by the E.W. 

 level, there occurred during the second year of observations 

 an extraordinary fall of the east end of the level, which it 

 has never since recovered. On Nov. 13, 1879, the inclination 

 was +1""66 towards the west, it fell gradually to — 5""80 

 on Nov. 26, and then the downward movement became rapid, 

 reaching — 88"*71 on Dec. 26. After this it rose slightly to 

 — 82"*16 on Jan. 5, 1880 ; but again fell, attaining its maxi- 

 mum lowering of — 89"°95 on Jan. 28. During the following 

 summer the east end rose no higher than — 74"'05 on Sept. 9. 

 M. Plantamour attributes this extraordinary movement in 

 part to the exceptionally low mean temperature of the winter 

 of 1879-80 ; but, as he observes, there must be some other 

 determining factor besides external temperature, whose cause 

 is as yet undiscovered. 



The following Table summarizes M. Plantamour's observa- 

 tions. In the last two columns I have added the average 

 inclinations of both levels throughout the whole yearf : — 



* These features are readily explained if the annual and diurnal 

 oscillations be due to the same cause. For then the resultant diurnal 

 movement would take place in the same direction as the resultant annual 

 movement at the same time, which is most frequently in an easterly, 

 and but rarely in a northerly, direction. I am unable, however, to test 

 this explanation, as M. Plantamour has only published the tables of 

 diui'nal motions for the E.W. level during the first year of observations. 



t A + sign before a figure indicates a rise, and a — sign a fall, of the 

 east and south ends of the two levels, with reference to their initial 

 positions on Oct. 1 and Dec. 23, 1878, respectively. From Oct. 7-21, 

 1879, no readings were taken. In calculating the average inclinations for 

 this year, 15 arithmetic means have been inserted between the figures for 

 Oct. 6 and 22, namely, +5"-85 and +5''-77 for the E.W. level, and 

 — 1"-70 and -2"-12 for the N.S. level. Excluding these 15 days the 

 average inclinations for the rest of the year would be —70 "•87 and — 2"*63. 

 It shoTild be added that in a few cases I have taken these figures from 

 MS. notes, not being able to re-consult the original papers. 



