( 431 ) 



l;ist coliiiim of llic iH-cccMliiiü' tahlc Tlici-c simmiis Io lx.' a slicht 

 ji'i-adiial ivlardalioii, l»iit lor llic rest llio reduced rales m llie main 

 a.<»'i-ee satisfju-loriiy, wliüe soiik> oj' ihe mosl diseonlaiil \ allies beloiiji- 

 lo short iulei-vais. As llie lengths of Ihese intervals are \ei-v dillereiil, 

 it would have little sense to derive a value for the mean dis- 

 ('oi'danee. 



4. TIk' I'Cihiciion of tin' iil)S(')T(ition.'< for /o/o/if/idc. For the 

 rediietion of m\' ol)ser\alions I followed the met hod jj,i\-en l>v Prof. 

 OlDKMANS '). 



I also follow e<l his advice not to reduce the transits to the middle 

 tliread. l»iit simply to take the mean of i\\o observed times of transit. 

 A sliuht disadxaida^e of the coiii-se followed l)\- me is that, when 

 either for the moon or for the star the transit o\er a thi'ead has 

 not been obsei-xed, we mns( also exclude the corres[)ondin,u- ti-ansil 

 of the other object. This was only the case with the two observa- 

 tions of 11)01 May 21 -). 



Having found in this n\ ay the times of the [)assage over the mean 

 of the threads we had now lirst lo determine the differences s l>etween 

 the observed differences in zenith distance of the moon and the star 

 as siiown by the level-i-eadings and the \alues computed for them 

 with adopted rediiclion elements, among w hicii an assumed longitude 

 of the station. Secondly the ecpiations of condition were formed, by 

 means of which the values of C are connected with the variatiojis of 

 the adopted elements. These (Mpiations were transformed so as to 

 express AA, the \ariatioJi of the ado[ited longitude, as a function 

 of ?, and of the \ariatioJis of the a<lopled latitude, of the adopted 

 cori-ection of the chronometer, of the times of transit of the moon 

 and the star, of the adopted right ascejision and deidination of the 

 two objects and of tiie ado[)le(l values for the parallax and the 

 semidiameter of the moon. 



On the one hand this course enabled me to correct the results 

 founil provisionally as soon as corrected xalues of the reduction 

 elements had been derixed: (ui the othei- hand the e([uations showed 

 the influence of residual errors in the obser\(Ml (piantities and in the 

 reduction elements. 



In the first place 1 shall consider the Naliies a<lo|>ted tinally for 

 the elements of i>'(liiction. 1 accepted as the latitude of the station 

 — 5M2'4".() (ContribiitioJis 1 p. (284) 11). The corrections of the 



^) Versl. en Med. Ki»n. Akad. Amslerdani tl, \Sol, p. ï25— iO. 

 -) By usinii- only curvespondinii- tiaiisits we also eliminate tlie itilluriicc ul the 

 refraction. 



