106 EIGHTH REPORT 1838. 



tion. When the district of observation is limited, the vari- 

 ation of this quantity may be disregarded.' 



The importance of an exact determination of these needle- 

 corrections is very great in the present instance. When, in- 

 deed, the same needle is employed throughout an entire series 

 of observations (as was done by Major Sabine in Scotland), it 

 is manifest that any error in the amount of its correction will 

 have the effect only of displacing the isoclinal lines in absolute 

 position, leaving their direction and interval unaltered. For 

 the dii'ection and interval of the lines depend solely on the dif- 

 ferences of dip ; and these are manifestly independent of the 

 correction, which alters all the dips by the same amount. The 

 case is different, however, when (as in the present instance) 

 different needles requiring correction are employed in the same 

 series. Here the differences of dip cannot be known, unless 

 we know the differences of the corrections of the needles em- 

 ployed ; and it is manifest that any error in the amount of that 

 difference will displace one entire group of results relatively to 

 the rest, and thus (when the mean geographical position of 

 these groups is different) induce a grave error in the direction 

 of the lines. 



Before we proceed to determine the amount of these errors 

 in the needles employed in the Irish survey, it may be desirable 

 to make a few remarks on their particular causes. 



Of the two sources of error above mentioned, the imperfec- 

 tion of axle appears to be the most common ; and it is to it we 

 are to ascribe (as Major Sabine has already remarked*) the 

 chief part of the discordances in the results obtained at West- 

 bourne Green in 1835. The same series, however, affords like- 

 wise a remarkable instance of the other error. Having pur- 

 posely destroyed the balance in two of my dipping needles, so 

 that they rested nearly in the horizontal position in Dublin, I 

 proceeded to use them exclusively for observations of intensity. 

 The results thus obtained were, however, so anomalous, that I 

 was compelled to reject them altogether. After some tedious 

 and vain attempts to discover the source of the anomaly, I was 

 at length satisfied, by a careful inspection of the results, that the 

 needles were under the influence of some other force besides 

 the earth's magnetism and gravity, and I concluded that this 

 disturbing force covild be no other than magnetism in the dip 

 circle itself. Trial soon verified this conjecture, and I had the 

 mortification to find that the apparatus which I had been so 

 long using was throughout magnetic, and that the magnetismf 



* Page 46. 



t Magnetism induced in ferruginous matter, not permanent. 



