20 Mr. Pratt's Second Letter on the Indian Arc. [No. 1. 



ing to the existence, or not, and the character of, other co-operating 

 disturbing causes. 



5. The great importance of this subject in the problem of the 

 Figure of the Earth — not the average figure of the whole earth, which 

 has been sufficiently well determined, but of the separate parts — 

 will be seen from the following facts : — 



I. — Colonel Everest, in his large and valuable volume of 1847, 

 assuming, as Mr. Tennant would do, that the Indian Arc is curved 

 like the average ellipse of the earth ; and ignoring, as Mr. Tennant 

 would do, the effect of the Himmalayas, brought out this result 

 (see p. clxxviii of his Introduction), that by his geodetic com- 

 putation Kuliana was farther north from Kulianpur by l-10th of 

 a mile than by the astronomical latitudes. He attributes this im- 

 portant discrepancy to mountain-attraction : but does not prove that 

 mountain-attraction will produce this exact amount of error. 

 My calculation shows that the effect of mouutaiu-attraction is, not 

 only to produce this amount, but a much greater amount, even 

 three times as great : and the only way of making things tally is to 

 assume that the form of the Indian meridian is not that of the 

 average ellipse. 



II. — If any one will turn to pp. lxx, lxxi of Colonel Everest's 

 volume he will see the great care with which the amplitudes* of the 

 arcs between Kaliana and Kulianpur, aud Kaliaupur and Damargida, 

 were observed. Eor example ; for the first, 36 separate stars were 

 observed and the average results taken. Of these separate 

 observations 29 differed from the mean by less than 1" in 

 excess or defect, 6 by less than 2" and one was 2"^ ; so near were 

 the individuals to the mean value ; and yet, to get an accu- 

 rate result, 36 observations were thought necessary— showing 

 that even a deviation of l" was considered of importance. But 

 Himmalayan attraction produces in this same amplitude (or differ- 

 ence of latitudes) an error of more than 15", and surely cannot be 

 passed by. 



III. — It is a fact palpable to the most ordinary observation, that 

 the surface of the earth is not that of a perfect spheroid (or ellipsoid 



* The amplitude of an arc of meridian is tho difference between the latitudes of 

 the two extremities of the arc. 



