E. L. Nichols — Age-coating in Incandescent Lamps. 277 



traneous source. The environment of the rock underwent a 

 third change and this was probably an elevation which strained 

 the albites, fissured the ottrelite and subjected the rock to nor- 

 mal surface weathering, during which the conversion of ottre- 

 lite to chlorite, was initiated. Prior to the granulation of the 

 clastic constituents the titanium in some combination must 

 have existed in the rock, but the mineralogical nature of this 

 combination is obscure. The most likely source of rutile is 

 from some titanium-bearing iron oxide the presence of which 

 has not been made out definitely except in the case of mica- 

 ceous ilmenite, itself manifestly of a secondary nature, occur- 

 ring as it does in a clastic rock, and which yields no evidence 

 of alteration. Ordinary granular ilmenite, such as occurs so 

 abundantly in phyllites, which is prone to decomposition, was 

 most likely the common source for all three minerals carrying 

 titanium, the rutile being an intermediate stage in the for- 

 mation of anatase. The micaceous ilmenite was developed 

 before the formation of the gneissic quartz, since the latter 

 encloses it ; the anatase probably forming after the quartz, as 

 it occurs in the interstices between the quartz grains. 



If this be a correct interpretation, the order of crystallization 

 of the existing minerals is essentially as follows : first, rutile and 

 micaceous ilmenite followed by the formation of gneissic quartz 

 enclosing them and coincidently the growth of sericite-enclos- 

 ing rutile. The glassy feldspars were then formed enclosing 

 all the previously formed minerals, and the anatase may have 

 resulted as an alteration product of the rutile at about this stage 

 in the rock's history. Then the ottrelite began its growth, 

 including all the other minerals in the rock and finally the 

 initial alteration of this mineral to chlorite closes its history up 

 to the present time. 



Cambridge, Mass., March, 1892. 



Art. XXXYI. — The Age-coating in Incandescent Lamps • 

 by Edward L. Nichols. 



[Experimental work by Messrs. B. E. Moore and C. J. Ling.] 



[Contributions from the Physical Laboratory of Cornell University, No. 11.] 



When" an incandescent lamp is maintained at constant volt- 

 age, it invariably falls off in candle-power. The diminution 

 in brightness is attended by an increase in the amount of energy 

 consumed per candle-power of light; this change being spe- 

 cially marked in the earlier portion of the life of the lamp. 



