324 MM. Strove and Dollen on the Correction 



certain about the moan density oi' the earth ; for there will 

 always remain the well-founded suspicion of some necessary 

 compensation. It is scarcely necessary to remark that all this 

 refers equally to those researches in which the intensity, in- 

 stead of the direction, of gravity has been chosen as the subject 

 of observation. At the same time it must be granted that the 

 danger of finding the observed difference disturbed by the 

 action of foreign causes is smaller the nearer the stations of 

 observation lie to one another. Accordingly we cannot quit 

 this subject without having shown in few words how by a 

 different procedure it might well be possible to set aside, if not 

 altogether, at least in great measure, the above objection, and 

 thereby to give to these observations an essentially different im- 

 portance. In our opinion the common failure of all these 

 researches lies in the fact that two stations of observation 

 only have always been thought sufficient, whereas it is really 

 by their frequency alone that a judgment can be formed as 

 to whether the disturbing mass which has been calculated for 

 is really all that has been effective*. The other remark to 

 which we referred above, has a more direct bearing on the 

 present subject. It seems that the second latitude, the deter- 

 mination of which we demand in order to justify a correction 

 being made to the first on account of local attraction, is so 

 essential that, while without it this correction has no foundation, 

 through it, in the greater number of cases, all further researches 

 become unnecessary. Let us examine somewhat more closely in 

 this connexion the results of the investigation into the deflec- 

 tion of the plumb-line at Schehallien. It was found that on the 



north side this deflection amounted to 11/ /# 61 x S, and on the 



o 



southern side to 9 7/ *17 x?;p and 8 betokening the mean densi- 

 ties of the mountain and of the earth. Now, as the difference 

 between the two observed latitudes proved to be ll"* 6 6 greater 

 than was demanded by the geodetic distance of the two points 



of observation, it was inferred that 20 //, 78x^=ll //, 66 ; or 



S= 0*561; whence the deflection on the northern side would 

 o 



* We must not ignore the fact that in the most recent determination of 

 this kind — at Arthur's Seat, near Edinburgh — there are three stations of 

 observation, one on the summit of the mountain, the other two on its 

 southern and northern slopes, almost in the meridian of its summit. This 

 seems to have been due, not so much to design as to the fortunate 

 circumstance that there a trigonometrical station occupies the summit, 

 and certainly not on account of any belief in what we consider to be the 

 material importance of a third station ; nor has its existence been taken 

 account of in deducing the result. 



