U�TIL comparatively recently, the determination of furnace and kiln temperatures by means of thermo couples was attended with complications and difficulties which made practical men impatient, and in some in stances doubtful in regard to the intrinsic value and reliability of the method. Technical men, however, ha\'e realized from the begin ning that, within certain temperature limitations, the re sults obtained by the use of thermo-couples were not only accurate, but could be made "direct

more »

... ng," pro vided that the thermo-couple be connected with a gal vanometer which would indicate the actual value of the e.mJ. obtained by heating. On first consideration, this problem did not seem dif ficult of solution, since it is an established fact that the e.mJ. or voltage of a thermo-couple bears a direct rela tion to the difference in temperature between its hot and cold ends. It is also true that the movable coil of a millivoltmeter will be deflected to an extent which is directly proportional to the e.mJ. of the thermo-couple. Therefore, it would seem the simplest operation imag inable to connect the thermo-couple and the instrument together by means of insulated conductors, figure the scale in temperature values, and regard the problem as solved. Unfortunately, however, we have also to contend with factors which cannot be ignored if accurate results are required; namely, line resist ance, thermo -couple resist ance, and the resistance of the indicator itself; because, to be precise, a direct current millivoltmeter actually is not a millivoltmeter at all, except in name, but is a milliammeter or current indicator.