240 Dr. BlAgden’s Report 
The fpecific gravity of ,825 having been fixed upon as the 
ftandard of re&ified fpirit in our tables, Mr. Gilpin was de- 
fired to afcertain by experiment what proportion of water 
would be neceflary, to reduce the lighteft alcohol in his poffef- 
fion to that ftandard. This was home alcohol, already men- 
tioned, which Mr. Lewis had furnilhed; and its fpecific gra- 
vity being ,814196 at 6o°, 3000 grains of it mixed with 135 
grains of diftilled water formed a compound, whofe fpecihc 
gravity was ,825153 ; that is, in round numbers, 100 grains 
of alcohol at ,814 with 4,5 grains of water, form our ftandard 
of fpirit at ,825. 
Perhaps fome perfons may object to the preceding experi- 
ments, on account of the fmall quantities mixed and weighed. 
Undoubtedly experiments on a large fcale have fome advan- 
tages ; but thefe in general depend more on the coarlenefs of 
tools, and clumfinefs of operators, than on the nature of the 
operations. If inftruments be exceedingly exa<ft, and the ex- 
perimenter equal to the talk of ufing them properly, I believe 
the errors upon moderate quantities will be quite as fmall in 
proportion as upon large; and in this particular inftance, where 
the greateft fource of error lay in the determination of the 
heat, the fmaller quantities had in that refpefl an evident 
advantage, it being much eafier to bring fix ounces of a liquor 
to an uniform temperature than fo many gallons. One of our 
moft eflential inftruments, namely, the balance, was fo much 
fuperior in nicety to any thing that could be wanted in thefe 
experiments, that error in weighing muft be thrown entirely 
out of the queftion. It was conftrudled by Mr. Ramsden * 
and fome account, though very imperfeft, of its admirable 
mechanifm, as well as of its extreme fenfibility, even when 
