447 
1881.J and its Effect on Astronomical Observations . 
angle gpc in Diagram 2 would be 2o"*7, and this 20 //, 7 is 
the amount that the polar distance of a star would be in 
error if seen from the Pole ; and twice 2 o"'y, that is 4i // *4, 
would be the difference in polar distance of a star found by 
two observations at intervals of six months. As this error 
would be nearly the same at both poles, it would be found 
that two stars near each pole would, when the sum of their 
declinations was taken at intervals of six months, differ to 
the amount of about 82". 
The question now arises whether any such differences 
have been found between any two stars when the sum of the 
declinations is taken ; and observations prove that this fadt 
exists, and has long been known, though a very different 
theory has been invented to account for the fadfs ; it having 
been assumed, on no evidence whatever, and in spite of the 
fadts shown by the unequal distribution of land and water 
on the Earth’s surface, that the axis of the Earth passes 
exadtly through the centre of gravity of the Earth, and that 
the poles of this axis are equidistant from all parts of the 
Equator. 
I will take, for example, the two stars 4 Draconis and 
ft Chamaeleontis, whose declinations are given in the 
“ Nautical Almanac of England and America.” These two 
stars are nearly on the same meridian of right ascension, 
and their declinations have been found as follows 
4 Draconis 78° 16' 30 // *4 N., Jan. 1, 1880. 
ft Chamaeleontis ... 78° 38' 33 ;/, 7 S.; Jan. 18. 
Sum 156° 55' 4"' 1 
In June these declinations are as follows : — • 
4 Draconis 78° iy' i ,/ *2 N. 
ft Chamaeleontis 78° 39' 22 // *5 S. 
Sum 156° 56' 2 3 // *7 
156 ° 55' 4"*i 
Difference T i9' /, 6 = 79 //, 6. 
Or exadtly the result which would occur in consequence of 
the centre of gravity of the Earth not coinciding with the 
Earth’s centre. 
Again, take the two stars Polaris and a Eridani, and, 
adopting the same principle, a difference of 75" is found in 
