1870.] 



Coimbra Monthly Magnetic Determinations. 



185 



ing the measures at the same nominal hour hy months, and taking their 

 monthly means for each hour. As these exhibited sensible discordance, 

 they were smoothed by taking the means of adjacent numbers, taking the 

 means of the adjacent numbers of the new series, and so on, repeating the 

 operation six times. The author explains the theory of this process, and 

 the way in which it tends to degrade the periodical terms of higher orders. 

 He then explains an easy method of resolving the numbers so smoothed 

 into periodical terms recurring once in the day, twice in the day, thrice in 

 the day, &c, and applies the method to the numbers for every month. 



When these quantities (which from month to month are perfectly inde- 

 pendent) are brought together in tables, they present such an agreement, 

 with gradual change accompanying the change of seasons, as leaves no 

 doubt on their representation of a real law of the diurnal changes of the 

 galvanic currents. They also show the existence of a constant turn 

 towards the north (which explains the apparent increase of force to the 

 north observed in the results for days of great disturbance), and a still 

 larger force towards the west (which also is well marked on the days of 

 great disturbance). No light is obtained on the origin of these terms, 

 but they appear to be probably pure galvanic accidents, depending on the 

 nature of the earth-connexions. 



The author then exhibits in curves the diurnal inequalities of mag- 

 netism which the galvanic currents must produce. The form generally 

 consists of two parallel lobes, making with the magnetic meridian an angle 

 of nearly 60° from the north towards the west. The greatest east-and- 

 west difference of ordinates, in the month of April, is 0*00044 of Total 

 Horizontal Magnetic Force ; it corresponds, in the hours to which those 

 ordinates relate, nearly with the Ordinary Diurnal Inequality. But it is 

 much smaller than the ordinary diurnal inequality, and the daily law of 

 the galvano-magnetic inequality differs greatly from that of diurnal in- 

 equality. For the greater part, therefore, of diurnal inequality the cause 

 is yet to be found. 



II. "Monthly Magnetic Determinations, from December 1866 to 

 May 1869 inclusive, made at the University of Coimbra/'' By 

 Professor J. A. de Souza, Director of the Observatory. Com- 

 municated by Balfour Stewart, F.R.S. Received December 

 16, 1869. 



[Note. — These observations form the continuation of a series the results 

 of which were communicated to 'the Royal Society on May 8, 1867, by the 

 President. In both series the same instruments were used, and the me- 

 thod of observation was the same in both. — B. S.] 



f 2 



