478 



Mr. J. E. PetaveL An JEzyjerimental 



perature of melting platinum, and it is difficult to predict at what tem- 

 perature they volatilise. 



Taking everything into consideration, it may therefore be said that 

 the present experiments confirm the theory that the crater of the arc 

 is at the temperature of volatilisation of carbon. 



The earliest determinations of the intrinsic brilliancy of the crater 

 were made in 1878 at Chatham under the direction of the Royal 

 Engineers' Committee.* The results varied from 39 to 442 candle 

 power per square millimetre, the mean value being 110. With regard 

 to more recent researches, Trotter gives the intrinsic brilliancy as 64, 

 Weber as 70, and Blondel as 158 candle power per square millimetre. 

 With the exception of those of Blondel all previous results are very 

 much lower than the values I have obtained. The discrepancy may 

 be attributed to the fact that instead of using a diaphragm to select 

 the rays from the centre of the crater, the first named observers 

 estimated the total area of the crater, and compared it with the total 

 amount of light. 



On the Lummer and Kurlbaum Incandescent Platinum Standard. 



In 1894 Drs. O. Lummer and F. Kurlbaum proposed a new standard 

 of light. t A strip of platinum foil 25 mm. wide, 0'015 mm. thick is 

 brought to incandescence by an electric current of about 80 amperes. 

 The temperature is increased until one-tenth of the total radiation is 

 transmitted through a water trough 2 cm. in width. This ratio is 

 determined by means of a bolometer. The construction of the instru- 

 ments require the greatest care, and three months had passed before I 

 was able to obtain the first observations. The instruments used for 

 this work are shown in fig. 4. The explanation appended to this 

 drawing is sufficient to render further description unnecessary. 



It would be useless to give details of the experiments which in the 

 main confirm the results already obtained by Lummer and Kurlbaum. 



With the same apparatus used in the same manner the light is prac- 

 tically constant as long as the ratio of radiations is adjusted to 1/10 

 per cent. 



The adjustment of the temperature of the platinum foil with the 

 required degree of precision is, however, most tedious, and in fact all 

 but impossible, except under the most favourable conditions. This 

 consideration, together with the complicated nature of the apparatus, 

 render this light impracticable as a working standard. 



At the beginning of the present paper it has been pointed out that a 



* R. E. Committee extracts for 1879 ; Report of the Electric Light Experiments 

 carried out at the School of Military Engineering at Chatham. 



f 'Berichte der Preuss. Akademie,' 1894, p. 227; ' Elektrotechnische Zeit- 

 schrift/ 1894, p. 475. 



