Dip-Needle Errors 



363 



fall near normal at this point could as well be drawn for one condition as for the other. 

 In this case the result is controlled by needles No. 1 and No. 5, which are chosen on the 

 evidence of adjacent stations. The complete analysis to determine the corrections 

 starting from observed needle-differences is given in Table 45. 



The second example, shown by Figure 5, illustrates a similar case from a region 

 of southerly or negative inclination. This short expedition comprises 21 stations in 

 South Australia, at which the observations were made by Observer A. L. Kennedy in 

 1914 using circle No. 41 with needles Nos. 1 and 2 of No. 178 and Nos. 5 and 6 of No. 41. 

 Of these needles the correction for No. 5 only had been determined by comparison, and 

 that at some time previous in a region of different inclination. Assuming the correction 

 thus determined to have remained constant, the corrections for the other 3 were worked 

 out by the foregoing method. The amplitude of the variation of needle No. 6 is greater 

 and takes place within a range of inclination less than has been found in other cases. 



-Correction-Curves for Needles used with Circle No. 41 during 1914. 

 grouped as indicated by plots for each needle.) 



(The 21 stations are 



