PROCEEDINGS OF THE POLYTECHNIC ASSOCIATION. 445 



cost nearly a dollar per foot. By directing a jet of oxygen 

 through an ordinary gas burner, the illuminating power of the 

 gas is greatly increased, and a saving of from forty to lifty per 

 cent, effected. The introduction of the oxygen into the flame has 

 also important consequences to health. It will destroy all the 

 noxious gases which have escaped the purifiers, and only water 

 and carbonic acid will result from the combustion. The amount 

 of these latter will be less than usual, for the reason that greater 

 illuminating eflect is produced by the employment of a smaller 

 quantity of gas. 



By the combustion of illuminating gas and oxygen nearly the 

 same heat is obtained as in the oxy-hydrogen blow-pipe. All 

 metals can be fused by this means if placed in suitable crucibles; 

 and the cost of large furnaces and expensive fuel will be saved in 

 numerous industries. 



The Chairman read the following paper on 



The Galvanometee. 



The name of this instrument imiDlies that it is used in measurino- 

 tlie force of an electric current. It has not, however, been made 

 entirely availaljle for that purpose, owing to the fact that the action 

 of the needle is not proportional to the force developed. For 

 very short distances, say fifteen degrees from the line of rest, the 

 arc described hy the needle may be said to be as the strength of 

 the current, because this arc and its tangent do not diiier materi- 

 ally in length ; but beyond that point the divergence of these 

 lines increase so very rapidly the eye loses their relation. It is 

 true that by means of the formula given many years ago by Am- 

 phere, with the requisite data, the graduation for a single instru- 

 ment might be made nearly correct ; still, as any change in the 

 elements used would change the graduation, and as the liability 

 to error in estimating the data is by no means inconsiderable, it 

 will be readily seen the adaptation and adoption of a theoretical 

 scale is impracticable. 



Several empyrical methods for forming a scale of intensity have 

 been devised. Those by Becquerel, Nobili, Melloni and Poggen- 

 dorff, will be briefly noticed. 



Becquerel and Nobili used a series of currents combined. They 

 cx3nnected the galvanometer with a thermo-pile, and successively 

 heated the alternate joints of one, two, three, four, five, &g., pairs 

 of rods, and as the strength of the current was then as one, two, 



