SOLID MEASUREMENT. 237 



All water meters more or less destroy the head of water; that is, make 

 a considerable difference between the velocities at which the water 

 enters the pipe and leaves it, and, consequently, where the stream is 

 used to develope power, any water meter is a very serious drawback, 

 but where no power is wanted from it, that is immaterial. 



This machine before us, which is called Siemens' alcoholimeter, is 

 one used to determine both the quantity of liquid that passes out of a 

 still, and also the quantity of alcohol that it contains, and it is used 

 largely in Russia. There are two distinct parts, first, a water meter of 

 tolerably simple construction, not unlike what I said before about gas 

 and water meters ; a copper disc revolves in a box upon a horizontal 

 shaft, with three vanes at equal angles, and the water comes in at the 

 centre, and drives always in one direction, delivering at the circum- 

 ference like a water wheel or vertical turbine. So far, it merely mea- 

 sures the quantity of liquid that passes, and the lower scale upon the 

 instrument measures the volume of liquid that flows out of the still. 

 That may be, of course, either wholly alcohol, or wholly water, or any 

 mixture of the two. To indicate the quantity of alcohol there is another 

 counter, and the way it is set to work is this: There is a weight at the 

 end of a lever, calculated either to represent a certain weight of pure 

 alcohol, or of proof spirit, in such a manner, that if there were no spirit 

 at all a second counter would always stand at zero; while if it were 

 entirely pure spirit, it would count the same as the lower one, which 

 shows the quantity of liquid passing. The spirit which passes is 

 generally considerably below proof. When pure water passes, this 

 upper counter is not affected, but when, on the contrary, pure alcohol 

 is passing, a bob comes down and pushes a second lever, with a 

 curved edge over, and this second lever regulates the difference 

 between the two counters. You may say that its edge ought to be a 

 straight line, and so it could be, but for the fact that one pint of water 

 and one pint of alcohol do not make a quart, but rather less, and 

 consequently there is a slight curvature in it. What happens in 

 the two receptacles is this. The liquid is weighed in one and 

 measured in another. One scale represents the measure, and the other 

 represents the joint effect of the weight and the measure; conse- 

 quently, one scale records the total quantity of liquid that passes, and 

 the other represents either the absolute quantity of pure alcohol 



