JSt>9.] Mr. Pratt's Second Letter on the Indian Arc. 25 



3. These calculations have been gone through agaiu lately by 

 a practised computer, working them under my direction by another 

 course. The result is, that while the first calculation (of 1855) 

 makes the deflections in the meridian at Kaliana, Kalianpur and 

 Damargida to be 27".853, 11" .968 and 6".909 ; this revised calcu- 

 lation makes them 27".943, 12".047 and 6".790. The differences 

 are too trifling to be of any moment ; and what variation there is 

 rather aggravates the effect. All the separate errors in the first 

 calculation, the aggregate of which has made this small discrepancy, 

 have been detected, so as to make the two calculations exactly to 

 tally. If there be, therefore, anything wrong in these results, 

 it must arise solely from the heights being wrongly taken, 

 or the density being wrongly assumed. The density used is 

 that of the comparatively small rock Schehallien, and must be 

 rather under the mark than over it ; as that rock is but a few 

 hundreds of feet high, whereas the most important part of the 

 Himmalaya mass is two miles high, and the lower parts must be 

 denser, rather than lighter, from the pressure of the superincumbent 

 weight. As to the heights, my own persuasion is, that, if anything, 

 the most important heights are taken too small rather than too 

 great.* But they are all written down in the six tables for in- 

 spection and criticism : and nothing can be easier than to point out 

 which are too small and which too large. Any information of this 

 kind forwarded to me I will immediately make use of, to correct the 

 results — a work which will cost no labour and take but little time. 



4. The calculations in the latter part of the paper of 1855 (after 

 para. 47, p. 87) are more laborious : and here some numerical 

 errors have crept in, one of which Mr. Tennant has pointed out, 

 for which I thanked him in my former letter to you. These errors 

 have no effect, however, upon the results of the paper. This part 

 has, moreover, received a revise in my communications sent home 

 last September. The result regarding the effect on the curvature 

 of the Indian Arc will be modified — increased or diminished — accord- 



* The Himnialayas rise to more than five miles. But the greatest height I 

 have taken in any one place falls a trifle short of two miles. The heights of beds 

 of rivers above the level have been taken wherever they could be found, and not 

 of the overhanging ridges and peaks. 



E 



