220 Mr. W. H. Preece. Behaviour of Glow-Lamps [Mar. 26, 



2. I used 60 Faure-Sellon-Volckmar cells freshly and fully charged 

 up at each series of experiments. The current through the filament 

 was measured by a specially constructed and calibrated Ayrton and 

 Perry direct-reading spring ammeter. The galvanometer in the 

 shunt circuit was a sensitive tangent galvanometer of the Post-office 

 pattern. The current through the filament was regulated by varying 

 the number of cells. The current and electromotive force through 

 the shunt aec were calculated by the following method: — The re- 

 sistance of the galvanometer Gr was 1070 ohms, and a variable 

 resistance R was inserted in its circuit. When the current from a 

 standard Daniell cell (l'Q7 volt) is sent through the galvanometer, 



1*07 



we get a tangent reading d, which, since jq^q = '001, is the deflec- 

 tion, or tangent reading corresponding to 1 milliampere of current. 

 The deflection (d x ) given by the shunt current was first read without 

 any resistance being inserted. 

 Hence 



C=-^l milliamperes. 



d 



Resistance was then added and a second reading d- 2 taken ; then the 

 resistance r of the shunt is obtained from the formula 



r=R . — Gr. 



cZ x — c7 2 



The electromotive force producing the current is 



E = Cr. 



The shunt circuit includes the rarefied space ec, and it is the resistance 

 of this space that we desire to know. 



All the observations were simultaneously made by different observers. 

 While one observed the behaviour of the lamp, another read the 

 current flowing through the filament, a third read the currents in the 

 shunt, a fourth varied the electromotive force, and a fifth recorded the 

 results. The electromotive force of the cells was carefully measured 

 before and after the experiments. No photometric measurements 

 were attempted. 



3. Experiment 1. — The connexions were made as shown in fig. 1. 

 The lamp (No. 4) was a short (75 mm.) filament lamp, with a pla- 

 tinum plate 30 mm. long and 5 mm. broad. 



The variation of the current and the increase in the resistance of 

 the filament towards the end of the experiment, together with the 

 behaviour of the shunt, are very noticeable. It is quite clear that 

 when the electromotive force attained 82 volts, a critical point was 



