238 New Form of Mountain or other Barometer. 
appropriate mark, the under surface of the float and of course the 
surface of the mercury beneath it in the cistern must also coin- 
cide with the zero of the graduation. 
I apprehend that this symmetrical position of the float sical 
the tube, tends to make such a coincidence of the zero mark and 
the mercury-surface more exact than is likely to be attained in 
any other arrangement; which places the float at a greater dis- 
tance, and entirely on one side of the tube. No single point of 
the mercury-surface, perhaps, is ever to be taken as precisely in 
the normal plane of the base to the atmospheric column equili- 
brating the barometric column: but this equilibrium is made by 
pressure of an infinite number of atmospheric columns whose ba- 
ses are in different horizontal planes, or, what is the same, by the 
pressure of an aggregate column whose base is as irregular as the 
cistern surface. A float, then, whose horizontal base extends 
equally all round the axis of equilibrium, may be supposed to 
present the fairest average of these irregularities and a general re- 
sultant of these several pressures. 
In any event, its exactitude may be taken as within any such 
methods as the estimation of the capacity of the cistern, an Eng- 
lish manner of construction ; or the ivory point used in some con- 
tinental barometers; or the minimum visibile, the most used but I 
think the most objectionable of all. 
8. The cistern is a porcelain or glass dish of suitable size ; 
which, when occasionally cleaned, admits of the application of 
heat to drive off all moisture. Wood is not favorable for such a 
purpose ; because it receives a very smooth surface only with dif- 
ficulty. It is besides hygrometric: and all varnish which might 
be used to remedy this disqualification brings another quite as bad 
—in the action, which the resinous components of such varnish 
are apt to have on the mercury. 
What has been so far described, contains all the parts necessary 
for establishing a stationary barometer. I suppose, of course, 
though I have not yet mentioned it, that the mercury is pure, 
that its specific gravity is ascertained, that the tube has been 
boiled, and that in immersing it in the cistern no substance has 
been placed in contact with the mercury, likely to act upon or 
soil it. Asa suitable implement for this purpose, I have figured 
in fig. 5 a tool, which Mr. Green applies in such cases. It is of 
iron, covered with clean undyed leather: the spheroidal pad at 
