E. L. Nichols — Age-coating in Incandescent Lamps. 279 



obtained by the two independent observers (Moore and Ling), 

 were in agreement, only the mean of the two sets is given in 

 the following tables. 



In order to ascertain precisely what part in the decadence of 

 the lamp is attributable to the coating, it was necessary to keep 

 careful watch of the continually shifting electrical and photo- 

 metric conditions during its entire life ; and in order to main- 

 tain in each lamp the precise conditions under which it had 

 been determined that it should be burned, measurements of 

 candle-power and voltage had to be made at short intervals, 

 together with continual readjustments of the current. 



Fourteen lamps in all were selected for the investigation. 

 These were of two widely different types, representing the two 

 great classes, viz : lamps with untreated and with treated fila- 

 ments. Of these, some were maintained at the voltage indica- 

 ted by the maker, or at predetermined, constant voltages other 

 than this. Others were kept at constant candle-power through 

 out their entire life, the current being increased whenever the 

 decrease in brightness due to increasing age had become 

 appreciable. 



From the data obtained in this way, the life curves of the 

 various lamps were plotted, showing the variation in candle- 

 power or voltage respectively, of resistance and of efficiency 

 expressed in watts per candle, during its entire existence. 

 Some of these curves have already been published.* As these 

 results were quite in accordance with those obtained by Mr. 

 Pierce, to which reference has already been made, it will be 

 necessary to deal with them here only in so far as they bear 

 upon the question of the influence of the age-coating upon the 

 efficiency of the lamp. 



Color of the coating. . 



Measurements upon all fourteen of the lamps which were 

 subjected to the spectro-photometric tests, showed the color of 

 the coatings upon lamp-bulbs containing treated and untreated 

 carbons, respectively, to be practically identical in character. 

 The color of the coatings obtained in lamps of either type, 

 under widely diverse conditions (as for example, by maintain- 

 ing a sixteen candle-power lamp at sixty-four candles through- 

 out its life), was the same as that produced under normal con- 

 ditions of service. The color in question is very nearly neutral, 

 the percentage of light transmitted being approximately the 

 same for all parts of the visible spectrum. The effect of the 



* The Artificial Light of the Future : Electric Club Pamphlets, No. 24, New 

 York, 1890; also Electrical World, 16, p. 387, Electrical Engineer, 10, p. 595. 



Am. Jour. Sci.— Third Series, Vol. XLIV, No. 262.— October, 1892. 

 19 



