114 



EIGHTH REPORT 1838. 



The following table contains the final mean dip at each sta- 

 tion, reduced to a common epoch, (the 1st Janliary, 1837,) ; and 

 the latitudes and longitudes of the stations : 



Table XXXVII. 



Of the foregoing results, those obtained at the Giants' Cause- 

 way and at Colerain are manifestly aflPected, to a very consider- 

 able extent, by the disturbing action of the basaltic rocks. The 

 effect of the basaltic pillars of the Causeway upon the magnetic 

 needle has been long since observed ; and on comparing the dip 

 recorded in the preceding table, with that due to the geogra- 

 phical position of the station, we find it in excess to the amount 

 of 50'. At Colerain, on the other hand, the effect of the 

 disturbing action has been to diminish the dip, but in a less 

 amount. The cause of these irregularities being apparent, we 

 have no hesitation in rejecting the results, in the computation 

 of the the isoclinal lines. 



Before we proceed to this computation, we must estimate the 

 weights of the observed results ; and for this purpose it is ne- 

 cessary to know the amount of the probable error of station. 

 This is obtained by computing (with assumed approximate 

 values of L,M,N,) the probable dip at each station, due to its 

 geographical position, and comparing it with that observed. 

 The sum of the squares of the diflferences of the computed and 

 observed results, substituted in (12), will give the total mean 

 probable error ; from which (the errors of observation and of 

 instrument being already known) the local error is deduced 

 by means of the equation (17). 



