28 On Magnetic Yariations. 



Diagram No. 1, was formed by projecting the whole of these 

 results, except the two first. It is at once evident that some of the 

 observations must have been in error, especially from 1813 to 

 1823, and again in 1841 and 1861. Between King and Oxley, 

 though only a year apart in time, the difierence is 18'; and 

 between Eowe and Sir Thomas Brisbane, there is a difference of 

 21', though only a year apart ; and there are two intermediate 

 results, both of which are probably good. In 1811 it is evident 

 that the assumed variation is the true one, and the same with 1861. 



A mere inspection of the curve from 1813 to 1823, combined 

 with the fact that a mean of the diiferent results then obtained 

 would put the curve just where we might expect, about 

 8° 50', would seem to indicate that the curve should be so drawn^ 

 but, I think it is better to retain the present curve, even though 

 it appears irregular ; for there are the satisfactory results of 

 five observers for its present position, and only two for changing 

 it — one of these had in all probability no delicate instrument to 

 determine the variation, and the other obtained three most un- 

 satisfactory results. 



At the present time (July, 1871) the variation at Sydney is 

 9° 35' East, and the same variation Avould be found all along the 

 line joining Sydney and Cooma. Lines drawn on the map of the 

 colony parallel to this, represent lines of equal variation through- 

 out their length ; approximately places one degree of longitude 

 from each other will differ 21' in variation, the vai-iatiou decreas- 

 ing to westward. 



In other words, the variation is 21' less for every fifty-eight 

 miles westward of Sydney. 



The greatest diurnal range of the magnet occurs in February,, 

 and amounts to 13', the magnet pointing 6' West of its mean 

 reading at 9 a.m., and 7> East at 2 p.m. 



As a general rule, 7 p.m. to 8 p.m. is the best time to take 

 the variation, or get the true magnetic medium, because at that 

 time the magnet is quieter and deviates less than at any other time 

 in the day. 



Diagrams 2 to 13. — I have represented the average direction 

 of the magnet at each hour of the day in each month of the year. 

 The curves are formed from the results of hourly observations,, 

 taken at Hobart Town from 1811 to 1818, and published in the 

 " Philosophical Society's Transactions for 1851" (paper by Lieut. 

 Colonel Edward Sabine, R.A.) From a number of hourly obser- 

 vations made here, I find the magnet follows a similar curve, and 

 I have given the results, because they will be useful to indicate 

 the best time to take the variation ; and in cases where the varia- 

 tion can only be observed at a particular time in the day, will 



