METHODS AND RESULTS OF TORONTO OBSERVATIONS. 75 



however, be safely asserted, that all the greater disturbances when 

 their mean effects are taken for a sufficient period of time, have a 

 character of periodicity. This was first suggested by General Sabine 

 in his comments on the observations at the Toronto Observatory for 

 the years 1841, 1842, and further comparisons have elicited the fact 

 that the disturbances (speaking of the declination) are more frequent 

 at the equinoxes than at the solstices, and occur most freqently at 

 night, the proportion of those occurring at night, to those in the day 

 being approximately ::8 : 5 These disturbances have also a regular 

 period approximately eleven years, a maximum occurring between 

 1848 and 1853, with 1843 and 1856 as the years of minimum dis- 

 turbance increasing again to a maximum in i860. Such regular 

 fluctuations in the amount of disturbance preclude the idea that 

 these differences are accidental. 



Additional evidence of the existence of this eleven year period 

 may be found in M. Arago's observations taken in the year 1821 to 

 1830, showing minimum daily range in 1823 and maximum in 1829, 

 1882-3 is a l so expected to be year of maximum disturbance, and 

 accordingly we have already had a disturbance far greater than has 

 been observed for many years. 



The progressive increase in the range of the diurnal variation 

 concurrently with an increase in the number and values of disturbed 

 observations at places widely apart, suggests the idea that they 

 proceed from some common cause, and though we cannot at present 

 say by what physical agency these disturbances are produced, still, 

 since we find that variations like the regular diurnal variations have 

 also a diurnal law, and since the sun is at least a primary source of 

 all magnetic variations which depend on local time, it is natural to 

 enquire whether the sun has any periodical variation having a coin- 

 cident epoch. 



Now, M. Schwabe's investigations, extending from 1826 

 onwards, seem to show an affection of the solar atmosphere, whose 

 periods of maxima and minima exactly coincide with those of the 

 magnetic disturances and the extent of the diurnal range of the 

 needle. In Humboldt's Cosmos will be found a table containing the 

 results of M. Schwabe's observations of sun spots from 1826 to 1850 

 and in Walker's treatise on terrestrial magnetism the same table 

 extended up to 1864, and from this we find that the following are the 

 minima years: 1833, 1843, 1856 and 1866. 



