POOR, THE FIGURE OF THE SUN 



413 



means of each group formed. When this is done, we find that the observa- 

 tions arrange themselves as in the following table: 



Date. 



P.-E. 



Weight. 



1880, June . 



1881, October 



1882, March . 

 1882, June . 



1882, November 



1883, May . 



1884, March . 



1885, January 



+ 0'M0 

 + 0".ll 

 + 0".06 

 + 0".05 

 + 0".05 

 — 0'M5 

 +0".28 

 — 0".17 



5.1 

 3.1 

 11.6 

 20.4 

 8.1 

 9.0 

 0.3 

 0.6 



We thus see that during the interval from 1881 to 1883 there is a pro- 

 gressive change; the equatorial diameter apparently growing longer in 

 relation to the polar diameter. While the division of the observations of 

 the year 1882 into three groups is more or less arbitrary, yet, no matter how 

 these observations had been grouped, the progressive character of the 

 change would have been apparent. The mean of all the determinations 

 for the year 1882 is + 0''.05 with a weight of 40.1. 



Tabulation of Results. — Collecting the results of the various 

 determinations into a single table, we have the following data from which 

 to determine the shape of the sun. In this table the measures are all reduced 

 to the same form, and give the excess in arc of the polar radius over the 

 equatorial. For convenience of comparison the series of Auwers and 

 Ambronn are separated into groups, the observations of various years or 

 series of years being collected together. 



Table XI. 



Meridian Observations. 



Date. Observer. P.-E. 



1809 Von Lindenau +5".0 



1831 Bianchi — 3".0 



1871 Secchi — C.S 



1885 Auwers — 0".2 



Heliometer Observations. 



Date. Observer. Series P.-E. 



1891 Auwers 1873-75 +0".05i 



1891 Auwers 1880-82 +0".03 \ 



1891 Auwers 1883-85 — 0".01 J 



1905 Ambronn 1890-91 +0".05-| 



1905 Ambronn 1893-94 — 0".02 • 



1905 Ambronn 1894-00 +0".01 . 



+ 0".019 



+ 0".013 



