480 Professor STEVELLY on a new Self-Registering Barometer. 
the pressure being maintained unchanged during each pair. Should the fluid in which 
the hydrometer is sustained, be any other than mercury, the division of both sides of 
the equation from which we deduce (_) by w, cannot be resorted to, as the w which 
multiplies the second member, is the weight of a cubic inch of that other fluid ; but 
practically this would be-of little consequence, as the form of the equation (L) would 
not be affected by this difference. 
As I stated in the beginning of this paper, it is obvious that a self-registering rain 
gauge may be constructed on a similar principle, by conducting the rain from a 
proper funnel into a vessel like the cistern sustained by a hydrometer, the descent of 
which would become an index of the quantity of rain received by the funnel, and a 
pencil might be made to trace at the several portions of the entire day, the height at 
which the gauge stood at each instant by means of a sheet of paper carried across it 
by clock work. 
