EEPOET OF THE SECRETARY. 19 



series of orclinar}' records. For this purpose deviations from the nor- 

 mal position of the bar amounting to tliirty -three scale divisions were 

 considered as due to abnormal disturbances of this kind. This num- 

 ber was ascertained by a theorem founded on the doctrine of proba- 

 bilities, for which we are indebted to Professor Pierce, of Cambridge. 

 Out of 24,231 observations, 1,G98, according to this criterion, were 

 considered as abnormal disturbances. After all the larger disturbances 

 were excluded, new monthly means were taken, and all deviations 

 from these of thirty-three divisions were again set apart. 



The result of this elaborate investigation is, that the variation in 

 intensity of the horizontal component of the earth's force is subject 

 to a change coincident with that fourid from observations in other 

 parts of the world, which has been called the eleven-year period, or 

 that corresponding with the frequency of the appearance of the spots 

 on the sun. The observations extended only through five years, 

 which was less than half the whole period ; the data were not there-' 

 fore sufficient to determine the movement through its complete cycle, 

 although they served to mark the minimum point and the ascending 

 and descending parts of the curves of illustration between two 

 maxima. 



Tiie fifth part relates to investigations on the effect of the sun in 

 producing daily and annual variations of the horizontal component 

 of the magnetic force. These investigations were made upon the 

 quantities which remained after removing from the tables the larger 

 or fitful disturbances previously described, leaving the normal effect 

 due to the sun from a change in its distance from the earth, and per- 

 haps from a change in the relative position of its magnetic poles on 

 account of the revolution of the earth in its orbit. Blanks in the 

 tables of observations were filled by interpolation. From these in- 

 vestigations it appears that the horizontal magnetic force of the earth, 

 or the intensity with which a deflected magnetic needle is drawn 

 back toward the north, varies with the different hours of the day. 

 The variation is greatest in summer and least in winter; the force is 

 most intense at a little after G o'clock in the morning, and diminishes 

 rapidly until a little after 10 o'clock a. m., when it reaches its mini- 

 mum, and again commences to increase, and continues to do so until 

 about 3 o'clock p. m., when it again declines until about 11 o'clock 

 at night. The amount of change is much greater in the day time 

 than at night, the whole indicating that this daily variation depends 

 on the varying heat derived from the sun. 



