ON A NEW CURRENT-METER. 169 



The zero-point of the sliding weight is naturally dependent 

 upon the density of the water, an increase in the latter increa- 

 sing the buoyancy of the pendulum-bob. If the instrument is 

 not adjusted for very great sensitiveness, only the salinity need 

 be considered, as the alteration of the zero by the temperature 

 of the water never exceeds some few miUimetres. 



If S be the salinity per mille, the zero, 0, of the sHding 

 weight, determined by the distance from the upper end of the 

 pendulum-rod to the upper edge of the sliding weight, is 



z = [12-9 + 0-133 *S] cm C 



For the augmented sliding weight with a total weight of 

 762 gr., the zero, also determined by the distance to the 

 upper edge of the weight, = in fresh water, and in salt 

 water 



z = 0-026 S cm. 



An explanation will be given later on of all these determina- 

 tions. 



We thus have in the three equations, Ä, B, and C, all that 

 is required to reduce all the observation-figures to one common 

 measure, viz. the pressure of the current in grammes upon the 

 bob and the wings of the pendulum. The relation between this 

 quantity and the velocity of the current, must be determined by 

 experiments. I have made temporary experiments such as 

 these with no great accuracy, and the result is represented 

 by the diagram, PI. XVI. 



It is thus easy from the observation-figures to find the velo- 

 city of the current with a sufficient degree of accuracy. I cannot 

 here give any definite opinion as to this accuracy, as the cur- 

 rents in the fjord where the apparatus was tested, were not 

 sufficiently regular for this purpose. I believe, however, that I 

 am safe in saying that for velocities varying between 1 and 

 15 cm. per second, they could be determined from the ice with 



