II .
" On the Deflection of the Plumb-line in India caused by the Attraction of the Himalaya Mountains and the elevated regions beyond , and its modification by the compensating effect of a Deficiency of Matter below the Mountain Mass .
" By the Venerable Archdeacon PRATT .
Communicated by Mr. STOKES , Sec. R.S. Received October 25 , 1858 .
( Abstract .
) The author begins by referring to his former paper , published in the 'Transactions ' for 1855 , in which he calculated the deflections caused by the mountain mass on the north of HIindostan , at three principal stations of the Great Arc , in the plane of the meridian , viz. Kaliana ( lat. 290 30 ' 48 " ) , Kalianpur ( 240 7 ' 11 " ) , and Damargida ( 18§ 3 ' 15 " ) .
He made them 27"'853 , 11"'968 , and 6"'909 ( or more correctly , as revised in the present Paper , 27"'978 , 12"'047 , and 6"790 ) ; and showed that the comparison of these two portions of the arc-which , if it be elliptical , and if the amplitudes are accurately known , ought to give the exact ellipticity of the arc in question-gives an ellipticity of L , instead of the mean -1- .
2 .
He next states that the Astronomer Royal , in a subsequent communication ( in 1855 ) , suggests that there is most probably a deficiency of matter immediately below the mountains which will cause a negative deflection , and so compensate for the mountain attraction .
Three objections are urged against this hypothesis , as stated by Mr. Airy .
It requires ( 1 ) that the solid crust should be only about ten miles thick ; ( 2 ) that the crust should be lighter than the lava on which it rests ; ( 3 ) that wherever there is a protuberance upwards in mountain masses and table-lands , there must be a corresponding projection of the crust downwards into the fluid , which it is difficult to conceive , as the same reason which is used to show it would prove also that , where there are hollows above as in deep seas , there must be corresponding hollows in the solid crust below filled up by the lava , and this would point out a law of varying thickness in the crust which no process of cooling could well produce .
3 .
The author considers , however , that if there be a compensating cause it must lie in this direction ; and he puts forth the hypothesis of deficiency of matter in a new form .
He supposes that the mountain mass has risen up in consequence of a slight expansion of the solid crust below through many miles of thickness , producing a slight attenuation from a considerable depth .
He calculates formulae and reduces them to tables to find the effect of this attenuation , and shows that the mountain attraction , modified by this attenuation , if it extend down through 100 or 300 , or 500 or 1000 miles ( the attenuation being uniform along each vertical line ) , will produce the following deflections : At Kaliana ... .
1"538 or 6'872 , or 10"8912 , or 16"-779 .
, , Kalianpur. .
0"`064 , 0"369 , , 2"`425 , , 4"'661 .
, Damargida 0"'065 , 0"076 , 0"120 , , 1"570 .
4 .
These four sets of deflections are then applied to correct the amplitudes of the two portions of the arc , and by their comparison to find the ellipticity , which is shown to be , in the four cases , 111d1 2i6 ' 280 ' 286 ' and 85 So that although the hypothesis , if the depth of attenuation be about 100 miles , greatly reduces the deflection , it does not reduce the ellipticity to the mean value , which is attained only if the depth be somewhere between 500 and 1000 miles .
There is little or no ground , therefore , for working with a mean ellipticity as is done in the Great Survey .
5 .
It is next pointed out that this theory will not explain the peculiarities of the Indian Arc under consideration ; in which ( according to Colonel Everest : see his volume for 1847 , p. clxxvii ) , the upper portion has an excess in its amplitude , geodetically determined , of 5"'236 , and the lower a defect of 3"`789 .
The presence of other disturbing causes near Kalianpur or Damargida , or both , is indicated by this ; either in visible masses above , which ought to be accurately surveyed ( as even small masses , if near enough , will produce the effect ) ; or in invisible defects or excesses of matter below , which it is impossible to discover , and therefore to estimate .
The possibility , and even not small probability of such existing without our being able to detect and estimate them , throws an air of doubt and uncertainty over all geodetic operations , whenever it is necessary to know with precision the position of the vertical , freed from the influence of local disturbing causes .
This is necessary for determining the curvature of the arc , that it may be used in both the problems of mapping the country with extreme accuracy , and of ascertaining the form of this part of the earth .
A note is appended , illustrating the degree of influence which errors in the verticals and the ellipticities may have on the mapping .
6 .
The author next applies the results of his paper to ascertain the effect upon the plumb-line of an excess or defect of density , of only 1-100th part above or below the density required by the fluidtheory of equilibrium , and prevailing over wide-spread spaces in the interior of the earth .
From the fact that specimens of rocks , even of the same description , found at the surface of the earth , vary considerably in density , he infers that it is not improbable that there may be as wide variations of density among the masses below , in addition to the variations arising from difference of distance from the centre of the earth and required by the fluid-theory of equilibrium .
If this be the case , his calculation shows that his fears expressed in the last paragraph are not unfounded .
The result of this part of the calculation is expressed in the following Table : TABLE OF DEFLECTIONS caused by a defect or excess of matter throughout a semicubic space of four millions of miles [ i. e. 200 each way parallel to the surface , and 100 miles in the vertical ] , the mean density of the excess or defect being 1-100th part of the density of the earth at the depth of the centre of the cubic space .
Depth of the Distance of the middle point of the space from the centre of the station , measured along the chord to the surface , semi-cubic space .
---379 miles .
581 miles .
781 miles .
980 miles .
1173 miles .
50 miles .
1-940 0'835 0'457 0-248 0'118 150 , , 1621 0-803 0-456 0-252 0-120 250 , 1-383 0-782 0-483 0-272 0'131 350 , , 1067 0'749 0-490 0-286 0-142 450 , , 0663 0-713 0-425 0-277 0'145 If the space be nearer to the station , or if the difference in density be more than 1-100th part , these deflections must be multiplied by a corresponding quantity .
7 .
The paper is concluded by a revision of some of the calculations in the former communication .
The mass of the mountain region abQve the level of the plains is shown to be somewhat more than four-millionths of the mass of the earth .

