126 



DR L. BECKER ON THE SOLAR SPECTRUM. 



The probable error of the estimations of intensity of blackness was derived from those 

 regions where no telluric lines were found. High and low sun observations were 

 treated separately. The number of instances was counted in which the estimated 

 intensity was the same as the final mean value expressed in whole numbers. The same 

 was done for all the observations that deviated ± 1, ± 2, &c, classes from the average. 

 The results are found in the following table : — 



Table III. 





High Sun. 



Low Sun. 



Region of Spectrum, \ in 10 ~ 6 mm 



566-552 



538-511 



497-486 



566-552 



538-511 



497-486 



Number of single observations, .... 



830 



1844 



710 



1501 



2855 



1006 



Number of observations in 100 of intensity equal mean 



71-5 



71-6 



72-8 



63-8 



61-7 



61-0 



,, ,, ,, equal mean ± 1 



28-0 



27-9 



26-6 



34-6 



36-6 



36-5 



,, ,, ,. equal mean ± 2 



0-5 



0-5 



0-6 



1-5 



1-5 



2-1 



,, ,, ,, equal mean ± 3 









o-i 



2 



0-4 



Probable error of one estimation of intensity, 



i'b-25 



±0-25 



i'b-24 



±0-32 



±0-33 



±0-35 



If it is borne in mind that the intensity-scale progresses by whole numbers, it is 

 evident that the high sun estimations are as accurate as the scale allows. The low sun 

 estimations are a little inferior to them in point of accuracy, as would be expected from 

 the great variations in the intensity of the continuous spectrum. 



We further compared the mean intensity as observed in low sun with that in high 

 sun observations. All the lines which were darker in low sun than in high sun were 

 counted, and the average difference between their intensities computed. The same 

 was repeated with those which appeared feebler in low sun, and with those of equal 

 intensities at both altitudes. The lines were divided into two classes, one comprising 

 those fainter than 8, and the other the darker lines. Close double lines were excluded 

 unless they had been separated on all occasions. The spectrum was divided into zones 

 to show the changes in the regions where telluric lines are numerous. 



Table IV. 



Difference of Intensities of Low and High Sun Observations. 



Region of Spectrum, 



'o 06 















"o oo 

 u A 















A in 10- G mm. 



1 8 

 S « 



n 



Ai 



n 



Ai 



n 



Ai 



Numb 

 lines 



n 



Ai 



11 



Ai 



n 



At 



566-552 



322 



30 



+ 0-5 



30 







40 



-05 



26 



40 



+ 0-3 



10 







50 



-0-5 



535-524 



258 



34 



+ 0-6 



33 







33 



-0-5 



38 



32 



+ 0-5 



27 







41 



-0-7 



524-512 



246 



33 



+ 0-5 



35 







31 



-0-5 



61 



54 



+ 0-4 



15 







31 



-0-6 



. 500-486 



231 



18 



+ 0-4 



27 







55 



-0-6 



50 



40 



+ 0-6 



20 







40 



-0-5 



"S | f 602-590 

 a. s | 590-578 



291 



82 



+ 2-6 



8 







10 



-1-1 



10 







40 







60 



-1-6 



361 



69 



+ 17 



12 







19 



-0-8 



4 



25 



+ 0-1 







75 



-0-4 



§ .g -j 578-566 



328 



72 



+ 1-8 



17 







11 



-0-6 



19 



16 



+ 0-4 



16 







68 



-07 



'§>J 552-535 



409 



48 



+ i-o 



20 







32 



-0-6J 



42 



33 



+ 0-5 



14 







52 



-0-6 



«3 1512-500 



258 



56 



+ 1-5 



15 







29 



-0-7 



53 



43 



+ 0-6 



25 







32 



-0-6 



