abstracts: physics 35 



Furthermore, the ordinary, commercial candlepower scale cannot well 

 be adapted to precision work because it is practically impossible to so 

 adjust the comparison lamp as to make the scale read the exact candle- 

 power directly, especially since at least six standards should be employed 

 in making the adjustment. 



At the Bureau of Standards these difficulties are overcome by automat- 

 ically recording the photometric settings by dots on a sheet of paper, or 

 chart, containing a special printed scale which is capable of being accu- 

 rately adjusted to the record. The chart is wrapped on a cylinder which 

 rests on the bench with its axis parallel to the photometer bar. By 

 pressing a key located on the photometer carriage a record dot is printed 

 at the correct position on the chart for each setting, as it is made, and as 

 the key opens, the cylinder, by means of an electrically driven secondary 

 clock, is advanced a short distance thus preventing any two dots from 

 coinciding and at the same time showing the order in which the readings 

 are made. A large number of settings on a given lamp are thus made 

 quickly and without prejudice and their mean, which corresponds to the 

 center of mass of the cluster of record dots, is readily and accurately 

 determined. 



The special scale is rectangular in form, the upper and lower sides being 

 graduated into linear candle-power scales corresponding, respectively, to 

 the extreme adjustments of the comparison lamp within which it is 

 desired to work. The points of division in these two linear scales are 

 joined by straight lines, thus producing within the rectangle an indefinite 

 number of linear candle-power scales, so that for any arbitrary adjust- 

 ment of the comparison lamp between the limits chosen there will be a 

 scale which will exactly fit the record printed by dots on the chart. The 

 proper one is easily determined by preliminary readings on the standards 

 and the values of the unknown lamps are read off to the nearest one- 

 hundredth of a candle. 



This scale is adaptable to all arrangements of the photometer and is 

 adaptable also to the measurement, without the use of rotating sectored 

 disks,* of lamps differing considerably from one another or from the 

 standards in candlepower; in which case, to make the record of all settings 

 fall on the chart, the various lamps when measured are placed, respec- 

 tively, according to their intensity, at different points along the photo- 

 meter bar. 



In each and every case the scale reads in candle-power directly, and if 

 the values be written on the chart when they are first read from the 

 scale it becomes a simple matter for a second reader to check the work of 

 the first entirely by inspection. G. W. M. 



