407 



the current annual amount of secular change takes place by equal 

 aliquot portions in every month, and even in every fortnight of the 

 year. The magnitude of the annual change of the Declination at 

 St. Helena, 8' (or more precisely 7'' 93 in each of the eight years in 

 which the observations were maintained), and the comparative tran- 

 quillity of the tropical regions in regard to magnetic disturbances, 

 were circumstances which rendered St. Helena a particularly eligible 

 locality for an investigation of this nature. The result has been, to 

 remove secular change altogether from the category of atmospheric 

 or thermic relations, with which, in the absence of a correct know- 

 ledge of the facts, it has frequently been erroneously associated ; and 

 to show conclusively that it is a phenomenon of far more systematic 

 order and regularity than has been generally apprehended (Proceed- 

 ings of the Royal Society, vol. vii. pp. 6775). 



It has thus been shown, that, in each and all of the branches of 

 inquiry for which the institution of the Colonial Observatories was 

 recommended, they have accomplished the objects which were con- 

 templated, and have in many respects exceeded the expectations on 

 which their recommendation was founded. Nor has the scope of 

 their performance been limited to a mere registry of the observa- 

 tions, or to their publication in a crude and undigested form. It was 

 well remarked by an authority of the greatest weight, when address- 

 ing the British Association on the occasion of the assembly of the 

 Magnetical and Meteorological Conference at Cambridge in 1845 

 (Herschel, Address, p. xxxv), that " A man may as well keep a 

 register of his dreams, as of the weather or any other set of daily 

 phenomena, if the spirit of grouping, combining, and eliciting results 

 be absent." To advance by the simple and straightforward path of 

 inductive inquiry, in a science such as terrestrial magnetism in which 

 a physical theory has yet to be sought, the endeavour must be made 

 " to grapple with the palpable phenomena, seeking means to reduce 

 their features to measurement ; the measurements to laws ; the laws 

 to higher generalizations ; and so, step by step, to advance to causes 

 and theories." The mere observational part is not, and ought never 

 to be, viewed as the fulfilment of the duties of institutions such as 

 magnetic observatories ; those duties ought always to be held to in- 

 clude (either on the part of the Directors of the Obse&vatories them- 



