22 



Art. 3— K. Sotome ; 



No. 9 



1915 



Jan. 20 

 21 

 21 

 26 

 27 

 28 

 29 



19 

 19 

 33 

 34 

 25 

 25 

 25 



d 

 0.079 



0.060 



0.020 



0.010 



0.040 



0.022 



0.021 



1 division = 1. "26 



24 

 20 

 10 

 42 

 20 

 48 

 52 



1 division = l."28 



These results show tliat the resistance ^^ of the level is not of 

 a negligiljle magnitude, so far as the degree of accuracy required in 

 the zenith telescoj^e observations is concerned, and we can take it 

 as established that this defect is a common property of the spirit 

 level, as we have confirmed its existence both from the theoretical 

 and the experimental standpoints. 



Hitherto we have been considering the motion of the l)ubl)le in 

 case the level itself is at rest. But as the motion is purely of a re- 

 lative nature, we can convert the above obtained result to the case 

 when the bubble is at rest and the level moves. 



Thus, when the level is put into slight movement from rest, 

 the bubble will accompany it, and will fail to show the real move- 

 ment of the level. If the movement of the level is less than or 

 equal to the resistance ^i, as above obtained, the bubble would 

 show no displacement relative to the level. 



Moreover, when the level continues to move further, the 

 bubble will follow it to a certain degree. The equation of motion 



