.:ttv now measuring on the Continent. 189 



the meridian between Greenwich 51° 28' 40", and Formentera 

 38° 39' 56", 11, of which the mean length of a degree is 



111115.08 met., the ellipticity deduced ^s ——. Such are 



the results of the comparison of the arc of the parallel in the 

 latitude of 45°, with arcs of the meridian which intersect the 

 parallel, and are measured on the same part of the globe with 

 itself. If we would now compare it with arcs of the meridian 

 measured in latitudes widely different from its own, — with 

 that of India, between the latitudes of 8° 9' 38",39 and 

 18" 3' 23",G4, of which the mean degree is 110628,60 met., 



we shall obtain by the same formula -rrr-r ; and if with that 

 •' 290,5 ' 



of Peru, of which the limits are 0° 2' 31" S. and 3° 4' 32" N., 

 and the mean degree 110582,10, we obtain-——-. 



® 294,7 



An attempt to investigate with precision the Ellipticity which 

 might best conform to the combination of this new element with 

 preceding determinations, would obviously be to anticipate 

 that period when the value of the degree itself shall be ascer- 

 tained with the additional accuracy, which the prolongation 

 of the arc to Orsowa, nearly doubling its present extent, will 

 produce : in the meantime M. Nicollet has assumed the 

 arithmetical mean between the four results which we have given 



above, , as, provisionally, the indication of the general 



Ellipticity of the globe, afforded by the combination. Thus 

 much we may at least infer, that as no one of the results 



approaches so small an ellipticity as-—, and as it would 



require an alteration in the length of the degree of the parallel 



of nearly 30 metres to produce —— as the mean result of 



the combinations, it is the decided tendency of this method 

 of inquiry, no less than that of the Pendulum experiments, 

 to be irreconcileable to that proportion between the earth's 

 axes, with which it was so lately supposed that all methods of 

 experimental inquiry were in accordance. We have now the 



result of the pendulum experiments -— ; of those meridian 

 arcs which have been most accurately measured 278?' ^^ 



