( 421 J 



Li(f -=: a -\- l> shi II -j- (• cos ii 



;is (loi-ixcd scparatciv from llic voai-s l(S95 — 9(S, ainl from 1899 1902 



were iiol in .u'ood ai>f('(Mii(Mil, and llial hclwceii llio Z- r) for 1 he same 

 value of // from holli |»erio<ls tlicrc exist disliiicl svslemalic dilfereiu'cs. 



Therefore 1 liaxc also sohcd riu-orously the («((iialioiis for llic 

 scpai-ale vears. As the Lff foi- tlie pcM-iod investigated hij Nkwco.mi} 

 are less aeenrate (thei-efoi'c Nkwcomi? himself used oid\ the mean 

 values foi- the ^vhole pei-iod) I have solved for this period not the 

 equations for the sepai'ate years i)ul those foi' ^I'oups of 2 oi' 3 vears. 



[n this way 1 obtained the followinu' values of it, h and c To 

 the c of 1S95 — 1902 the corrections mentioned heforc^ lia\-e heen 

 ap[)lied. The values of <i dei-ived from the two sei'ies are not directh' 

 comparable inter se, as the declinations were not reduced to the 

 same svKstem ^). 



To these results we may add those for J<S92 dci'ixed by Fr.vnz. 

 I had at (irst o\erlooked the fact that Fr.vnz had not oidy discussed 

 his own obserx'ations of the ci'aler M(")stin,U" A, but also the I'esnits 

 of a similar series of obsei-valions made in the same \ears at 



1) Gomp. also the remark at the end of this section. 



