1245 
the same time care must however be taken that in the lower 
part of their path the balls have the same direction as the water 
current in the middle of the tube. The distance between the copper 
disk and the glass tube was made very small. A strong light- 
beam is concentrated on the lower part of the disk, so that the 
clear luminous points of the moving balls can be observed in the 
rotating mirror. The velocity of the copper disk is thus regulated 
that it is approximately equal to the velocity of the water in the 
tube. The observations on the disk with the balls are so much more 
accurate than those on the water-current that the correction by which 
observations in an inclined direction must be reduced to a plane 
perpendicular to the mirror becomes exactly known. This correction 
approximately amounts to 1°/, of the angle. 
6. The result of the experiment for a part of the tube at a 
distance of about 23 em from the point denoted by J” is given in 
the following table. 
V Jh ve ct a | om |p= ONCE 
3001 2623 | 330.3 | 44.9 | 443 | 389.9 | 0.847 
1051 985 | 308.1 | 46.2 | 45.6 | 372.6 | 0.827 
1609 | 1479 | 314.1 | 46.2 | 45.6°| 372.6 | 0.843 
1286 | 1175 | 316.0 | 45.6 | 45.0 | 380.5 | 0.830 
1617 | 1501 | 311.0 | 46.0 | 45.4 | 375.2 | 0.829 
| peas 
0.835 
Under J” is given the water-volume, expressed in liters, which 
has flowed through the tube in 7’ seconds. This amount has been 
corrected for the error (—1,6°/,) of the ‘Ster’meter, as it was 
given to me by the Direction of the municipal water-works. 
Then wv, foliows (in cm./sec.) from the transverse section of the 
glass tube. The angle @’ is half the directly read angle, « the value 
corrected for the inclination. v,—=v/tange@ gives the maximal velo- 
4al Ar a IRS on cm 
= —_ = 380,5 — is the velocity 
1.052 1,052 sec : 
which by means of the rotating mirror is added to the velocity of 
the water, where / is the ‘effective’ distance from the axis of the 
mirror to that of the tube. 
city in cm./sec.; v 
