( 451 ) 



or 190(1 Imt after liaving eom|»lete(l som-^ four years tliere seemed 

 reason lo iliink ihal lliere was hardly need for fiii-tlier information. 

 Tlie geiieral result arrived at was, that the tables were still sulKi- 

 ciently accurate for our purpose, which was no other than to prepare 

 astronomers for the observation of the mutual occultations and eclipses 

 of the satellites. 



Now that the work is finished w^e will not suppress its results 

 though it cannot at all claim to be complete. It never was om- 

 intention to make it so, and the journals appearing in France, in 

 America etc. have not been searched. 



The following observatories have contributed to our investigation. 



Aperture of the telescopes 

 in ni.ni. 



Greenwich 102, 170, 254, 714. 



Utrecht 260 



Uccle 150 



Jena (Winkler) 162 



Halifax (Gledhill) 237 



Pola 162 



Christiania 74, 190 



Kasan 66, 81, 84, 96, 244 



Göttingen 161 



Windsor (Tebbutt) near Adelaide 203 



Lyon (a single observation) 2 



At Greenwich, Cliristiania and Kasan the eclipses have been often 

 observed by two or more astronomers using telescopes of different 

 aperture. In such cases we have only taken into account the instant 

 observed by means of the telescope of largest aperture. As a rule 

 the observer at this telescope could follow the satellite longer at 

 "disappearance" and he would pick it up earlier at "reappearance". 

 There are however a few exceptions to the rule. 



For the eclipses observed during the period of a single opposition 

 of Jupiter the corrections to the data of the Nautical Almanac in 

 no case showed a regular progression. They fluctuated on both sides 

 of the mean in such a way that there could be no objection to 

 adopting their arithmetical mean, a proceeding wdiich still would be 

 pei'feclly justified, even if there had been a regularly increasing or 

 decreasing progression. No further attention was paid to the diffe- 

 rences in the aperture of the telescopes. If these apertures exceed 

 a certain amount, for instance 150 mm. we find, theoretically as 



