PYROMETER into a vessel containing snow and water, the balance of resistance between the two battery circuits would be obtained without adding va- riable resistance to the coil of constant resis- tance, and the needle of the differential gal- vanometer would remain at zero when the current is established. But on exposing the pyrometer to an elevated temperature, the re- sistance of its platinum coil would be increased, and resistance to the same amount would have to be added to the constant resistance of the measuring instrument, in order to reestablish the electrical balance. This additional resis- tance would be the measure of the increase of temperature, if only the ratio in which plati- num wire increases in electrical resistance with temperature is once for all established. This is a question which I shall revert to after hav- ing completed the description of the pyrome- tric instrument. Although I have stated that by means of a differential galvanometer and a variable resistance (constituting in effect a Wheatstone bridge arrangement) the increas- ing resistance of the platinum spiral may be measured, it was found that the use of a deli- cate galvanometer is attended with consider- able practical difficulty in iron works and oth- er rough places where it is important to meas- ure elevated temperatures, or on board ship for measuring deep-sea temperatures. I was therefore induced to seek the same result by the conception of an instrument which is inde- pendent in its action from tremulous motion, or from magnetic disturbance caused by mov- ing masses of iron, and which require no care- ful adjustment or special skill on the part of the operator. This instrument is represented in fig. 5, and may be termed a chemical resistance measurer or 'differential voltameter.' Fara- day has proved that the decomposition of wa- ter in a voltameter, expressed by the volume of gases V, is proportionate in the unit of time to the intensity I of the decomposing current, rr or that 1=^ According to Ohm's general law, the intensity I is governed by the electro- motive force E, and inversely by the resistance R, or it is 1=^. It is therefore ^=T[, or ET V=-jj-; or the volume V would give a cor- rect measure of the electrical resistance R, if only the electromotive force E and time T were known and constant quantities. But the electromotive force of a battery is very vari- able ; it is influenced by polarization of the electrodes, by temperature, and by the strength and purity of the acid employed. The volume of gases obtained is influenced, moreover, by the atmospheric pressure, and it is extremely difficult to make time observations correctly. It occurred to me, however, that these uncer- tain elements might be entirely eliminated in combining two similar voltameters in such a manner that the current of the same battery was divided between the two, the one branch comprising tie unknown resistance to be meas- ured, and the other a known and constant resistance. The volume of gas V, produced in this second voltameter, having a resistance R*V R, in circuit, would be expressed by V t =- t and we should have the proportion of V : V,=
- -^ ; but E and T, being the same in both
K K cases, may be struck out, and the expression will -assume the simple form V : V, : : R : 11,. The constant resistance R of the one circuit being known, it follows that the unknown re- T>y sistance R, is expressed by - - ; that is to say, by a constant multiplied by the proportion of gas produced in the two voltameters irrespec- tive of time, or strength of battery, or tem- perature, or the state of the barometer. The resistances R and R, are composed each of two resistances, namely, that of the principal coils, which we may term R or R /? and of the voltameter and leading wires, which is the same in both cases, and may be expressed by y. The expression should therefore be written as follows : V : V, = R, + y t : R -f y, R, being the unknown quantity. The mechan- ical arrangement of the instrument will be un- derstood from the diagram, fig. 5; and the Fio. 5. Siemens's Pyrometer, Resistance Measurer. whole arrangement of the pyrometer, with its leading wire and resistance measurer, from the general view given in fig. 6. The voltaic re- sistance measurer, fig. 5, consists of two cali- brated vertical tubes of glass of about three millimetres diameter, which are fixed upon a scale showing arbitrary but equal divisions. The upper ends of the tubes are closed by small cushions of India rubber pressed down upon
