152 PROCEEDINGS OF THE AMERICAN ACADEMY 



pheric absorption ; for while the inclination of the ecliptic changes by- 

 seven degrees in passing from each group of observations to the next, 

 the change in the final column of Table L is 0.30 between Groups 1 

 and 2, and only 0.06 between Groups 2 and 3. The corresponding 

 ehanjies in the latitude of the axis are 1°.9 and 0°.0. The evidence 

 of Weber, taken without regard to the number of observations in each 

 group, would not support the hypothesis of regular change ; but the 

 number of observations in his first group is only six, and this circum- 

 stance leaves us without any distinct result in his case. The mean 

 result for Heis and Weber, with equal weights for the separate ob- 

 servations, would indicate a progressive change in the latitude of the 

 axis. 



The small number of the observations of Eylert and Neumayer 

 makes the result from them inconclusive. It is of interest, because 

 we have no other observations made in southern latitudes except those 

 of Jones, and it tends to confirm the principle established by that 

 observer. But in this instance a change in terrestrial latitude ac- 

 companies that of the position of the ecliptic. 



Schmidt's observations, if they were sufficiently numerous, would 

 confirm Serpieri's hypothesis. The axis remains at two degrees of 

 north latitude until the observer changes his station to Italy, when it 

 begins to approach the ecliptic. But neither of the four groups con- 

 tains many observations, and there are some peculiarities in Schmidt's 

 results, to be considered hereafter, which may partially account for 

 the discordance between them and some of the others. 



Setting aside, for the present, the consideration of atmospheric 

 absorption, my conclusions from the observations thus far considered 

 are, first, that the experience of Jones with regard to the shifting of 

 the light is confirmed by that of other observers ; secondly, that he 

 was right in connecting the change with the altered position of the 

 ecliptic in the visible hemispliere, rather than with the change in 

 terrestrial latitude from one place of observation to another. 



In the absence of the strict reduction mentioned above (p. 149), 

 only a very general conclusion can be drawn from the observations at 

 the elongations 30° and 90°, since their number is often comparatively 

 small. It will appear by inspection of Tables XXV., XXVIII., 

 XXXI., and XXXIV. that the observations of Jones, Ileis, and 

 Welier, made at these elorgations, agree on the whole with the results 

 obtained at the elongation G0°. The other observers furnish very 

 little material for a conclusion, and the results are irregular. 



We have still to consider the observations by Jones of what he 



