Hydrometrical Observations. 385 
has been out of use for some time, before being again employed in mak- 
ing observations. A scale sufficiently accurate for most hydrometrical 
purposes (though not for the instrument when used as an anemometer) 
may be obtained by applying it to some regular channel, such as a mill- 
lead formed of masonry, timber, or iron, where the velocity is nearly the 
same throughout, and noting the number of revolutions performed du- 
ring the passage of a float over a given number of fect, measured on the 
bank. In this way it was found, by the mean of 62 observations, that 
each revolution of the vane in the instrument of which a drawing has 
been given, indicated the passage of the water over 46 inches. The 
number of revolutions at several parts of the stream was ascertained to 
be the same in equal times, at both the commencement and the end of 
the experiments. This number, therefore, becomes in the instrument 
alluded to, a constant multiplier of the number of revolutions indicated 
by the vane ; and hence, the number of feet passed over by the water in 
the given interval of time is ascertained. 
The direction of the under current, which it is sometimes interesting 
to know, cannot, however, be obtained by means of the tachometer, and 
I shall describe a plan for obtaining an approximation to both the velo- 
city and direction of under currents, which is of easy application, and 
may be useful to those employed in engineering investigations. The 
plan to which I allude was devised and used at the Cromarty Frith in 
1837, by Mr Alan Stevenson, who discovered, by means of the instru- 
ment he employed, the interesting fact, that, at the depth of 50 feet, the 
velocity of the current, at both flood and ebb, is in certain places of the 
Frith nearly double that at the surface. This instrument, which of course 
merely gave an approximate result, consisted (as shewn in the accom- 
panying cut, at letter a) of a flat plate of sheet iron, measuring 12 by 18 
inches, having a vane made of the same material, and measuring 4 feet 
in length, fixed at right angles to the centre of it. The lower edges of 
the plate and vane were loaded with bars of iron, for the purpose of 
causing the instrument to sink to the requisite depth ; and it was so slung 
as to preserve the surface of the plate in a vertical plane. This Appara- 
