352 



L. Bell — Absolute Wave-length of Light. 



In this case the error introduced by an error of 1" in <p is a 

 little less than 1 part in 250000. 



The mean value of 6 during these measurements was 



(9 = 6° 59' 58"'6. 



In case of grating IV the line selected for observation was one 

 of Rowland's standards at w.l. 5914 - 319 of his preliminary list. 

 It is a very close double, the components being distant from each 

 other something like 7-r.wo" °^ f ne i p wave-length. The double 

 deflection was as before 72° but in the fifth, order. As with 

 grating III eighteen series of observations were obtained, with 

 the following resulting values of f 



Dec. 



16, 



a 



16, 



u 



16, 



« 



19, 



a 



19, 



it 



19, 



Jan. 



12, 



u 



12, 



i< 



12, 



cc 



14, 



u 



14, 



u 



14, 



a 



19, 



u 



19, 



a 



19, 



a 



20, 



a 



20, 



« 



20, 



36° 



0' 



+i" 



•16 



36 







+ 



•66 



36 







+ 



•67 



36 







+ 



•64 



36 







+ 1 



•56 



36 







+ 



•85 



36 







— 1 



•19 



36 







— 1 



•61 



36 







— 1 



•79 



36 







— 1 



•09 



36 







— 



•95 



36 







— 



•89 



36 







— 



•48 



36 







— 



•59 



36 







— 



•49 



36 







+ 



•51 



36 







— 



•11 



36 







+ 



•55 



The mean value, corrected as before for error of thermometer, 



is : 



<p = 35° 59' 59 // -06 + // '15 



The effect of this probable error is obviously the same as in 

 case of grating III. The mean value of the semiangle between 

 the telescopes was 



6=6° 58' 31". 



During the observations with grating III the barometric 

 height reduced to the place of observation was very nearly 

 762 mm , but during the work with grating IV it was phenomenal- 

 ly high, reaching an average value of 766 mm , an amount so far 

 from normal pressure as to render a small correction necessary. 



The mean temperature during the observations with III was 

 about 21° C, but in case of IV it averaged almost exactly 20° C. 

 varying at most only two or three degrees from that figure. 



