THROUGH FINLAND. 381 
“ differ fenfibly from that of a fphere, the equation 11 — 0 will 
(i neceflarily contain one or more of thofe fmall quantities a, b, 
“ &c fo affefted, that if we fuppofe them =z 0, the remaining 
equation u — 0 will be that of a fphere, whofe dinyenfions are 
“ equal to thofe of the earth; whence it will follow that u will 
“ be the firft term of our developed feries, as w r e have juft faid. 
s ‘ Befides, if we except aA ^ , and bAo.l, all the other terms 
u a™b n A may be difregarded on account of the fmallnefs of 
m,n J ° 
“ the quantities a, b , &rc. ; confequently the furface of our globe 
will be reprefented by the equation U x q X bA ^ —: 0. 
“ The whole then comes to this, to know what is the form of 
the functions A and A q which being fuppofed of the fe- 
cond order, the equation u —■ 0 will reprefent an ellipfoid, 
“ whofe eccentricities of the equator, and of the meridian which 
“ paftes through the great axis of the equator, will depend on the 
“ quantities a, b , &c. In refpecl of the figure of revolution, it 
“ is very clear that then the quantity a will be nearly equal to b, 
“ or what is the fame thing, if we make a — b x b, the quan- 
(t tity u will be very fmall; whence it follows that the equation 
c< u — 0 , may be reprefented by this 0 — U X bA^ ^ . In fliort, 
“ there is no reafon to fear that the difregarded quantities fhould 
u ever prove confidcrable enough to produce any error of confe- 
(< quence, as difficulties wffiich in all probability we ffiall never be 
“ able fully to furmount, will for ever prevent our precifely know- 
(e ing 
