20 Captain Sabine's Experiments on the 



1823. April. Horticultural Garden, Chiswick. Th. 42°. Arcs 30° to 

 10°. Rate 164".0 + 0".5 to Th. 60°+l".08 for Arcs = 165".6. 



1828. March. In the Regent's Park. Th. 40°. Arcs 30° to 5°. Rate 

 1 63".52 + 0".56 + 0".7 1 = 1 64".8. 



1828. August. Horticultural Garden, Chiswick. Th. 64°.5. Arcs 30° 



to 5°. Rate 166".9-0".12 + 0".71 = 167".5. 



1829. July. Horticultural Garden, Chiswick. Th. 67°. Arcs 30° to 



5°. Rate 166".6 - 0".22 + 0".71 = 167".l. 



1821. October. Regent's Park 166".4j 



1823. April. Chiswick 165".6j 166,0 



1828. March. Regent's Park 164".8 > 



1828. August. Chiswick 167".5 tl66.5 



1829. July. Chiswick 167".l J 



Mean . 166".3 



In the case of this needle, we may consider the differences 

 from the mean rate as accidental deviations, occasioned pos- 

 sibly, in part, by the observations having been made at different 

 periods of the year, or at different hours of the day, or at dif- 

 ferent spots, and in part, also, by the errors of observation, 

 rather than as systematic alterations in the magnetism of the 

 needle; and we may therefore regard 166" as the rate in Lon- 

 don, corresponding to observations made elsewhere in the years 

 1822 and 1823. 



I have not attempted to introduce a correction for the period 

 of the year at which the observations were made at the different 

 stations, because I do not feel assured that we have sufficient 

 evidence of the amount of the change which the magnetic force 

 undergoes, in England, at different seasons of the year, inde- 

 pendently of the effect on the needle itself of variations of 

 temperature : and we have no satisfactory evidence whatsoever 

 in regard to the monthly oscillations of the force, at any of the 

 foreign stations. With respect to the hour of the day, all the 

 observations, at every station, were made between the hours of 

 noon and five P. M., excepting at Jamaica and Maranham, 

 where, from accidental circumstances, they were necessarily 

 made between six and nine A. M. No order of succession 

 was preserved in making the observations, the needle that first 

 presented itself being the first used : but any irregularities 

 arising from this cause disappear in the mean result. 



With needle No. 4, we have in 



