Royal Society* 491 



till it was pure, and was then treated with oil of vitriol and sea- 

 salt, it did not yield arsenic by distillation. In conducting this 

 process, the King^s yellow requires to be washed for a long time 

 before the whole of the arsenious acid is removed. In the trial 

 above referred to, boiling water was used repeatedly for one day, 

 and cold water was allowed to flow on it, in a constant stream, 

 for another day, before the water came off pure. The first wash- 

 ings were alkaline, and gave a yellow precipitate on the addition 

 of muriatic acid, and also by the transmission of sulphuretted 

 hydrogen, showing that it contained not only arsenious acid, but 

 also sulphuret of arsenic, held in solution by an alkali. 



King's College, Aberdeen, July 1851. 



LXXI V. Proceedings of Learned Societies, 



ROYAL SOCIETY. 



[Continued from p. 320.] 



May 22, *'/^N the Annual Variation of the Magnetic Declination, 

 1851. ^^ at different periods of the Day." By Lieut.-Col. 

 Sabine, R.A., V.P. and Treas. R.S. &c. 



In this communication the author has arranged and presented 

 together the Annual variations which the magnetic declination 

 undergoes at every hour of the day at the four Colonial Observatories 

 established by the British government, at Toronto, Hobarton, the 

 Cape of Good Hope and St. Helena. This has been done by 

 means of a graphical representation, in which the annual variations 

 at every hour are shown by vertical lines varying in length accord- 

 ing to the amount of the range of the annual variation at each 

 hour; each line having also small cross lines marking the mean 

 positions of the several months in the annual range. The mean de- 

 clination in the year at the respective hours is marked by a hori- 

 zontal line which crosse.4 all the verticals at each station. The 

 hours are those of mean solar time at each station, the day com- 

 mencing at noon. The annual variations represented in the plate 

 were obtained at Toronto from three years of observation, viz. 1845, 

 46, 47 ; at Hobarton from five years, viz. July 2nd, 1843 to July 1, 

 1848; at the Cape of Good Hope from five years, viz. July 2nd, 

 1841 to July ist, 1846 ; and at St. Helena from three years, viz. 

 July 2nd, 1844 to July 1st, 1847. 



The author observes that it is perceived at the first glance at the 

 plate, that the range of variation at all the four stations is consider- 

 ably greater during the hours of the day than during those of the 

 night ; and that there is a great similarity, though not a perfect 

 identity, at all the stations in the relative amount of the range at 

 different hours. Further, that the amount does not progressively 

 enlarge to a maximum at or about noon, when the sun's altitude is 

 greatest ; or at the early hours of the afternoon, when the tempera- 



