430 Prof. KreiPs Deductions, fyc. 



after the great oscillations have ceased, and only gradually 

 resumes its ordinary force. 



III. All the alterations of this element are accompanied 

 by changes of the dip, and may chiefly be ascribed to them ; 

 an increase of dip, and a diminution of the horizontal inten- 

 sity always taking place together, and vice versa. 



IV. The vibration of the dipping-needle is more rapid 

 during disturbances, consequently the total force is increa- 

 sed* ; and as we have seen that the horizontal intensity is 

 weakened at such times, the influence of the increase of dip 

 preponderates over the augmentation of the total intensity. 

 In countries where the dip is much less than at Prague this 

 may not always be the case, and the horizontal intensity may 

 increase during a disturbance. 



V. The variations of the total intensity frequently occur 

 simultaneously with those of the dip and horizontal force. In 

 half the number of perturbations which were observed con- 

 tinuously, the strongest total intensity coincided with the 

 highest dip and least horizontal intensity, or the weakest total 

 intensity with the smallest dip and greatest horizontal inten- 

 sity ; affording a further evidence that the variations of the 

 horizontal component are chiefly due to the changes of dip. 



VI. In a strong disturbance all the three elements were 

 usually affected, but in a variable degree, which may probably 



* This is in contradiction to the 23rd paragraph in my letter to M. 

 KupfFer, (Phil. Mag., April 1840, p. 249), and I hope that it may prove a 

 rectification. I consider the method applied to the Milan observations an 

 unsafe one, as by it the daily means of the times of vibration on days of dis- 

 turbance were compared with the average time of the whole month. Such 

 a comparison cannot show with exactness the small effect which the pertur- 

 bations produce in this element, because in the course of a month it is sub- 

 ject to too great changes, whether real or instrumental. The greater fre- 

 quency of the observations in each day at Prague, enabled me to proceed in a 

 different manner, namely by comparing the mean of a disturbed day with those 

 of the preceding and following days. Occasionally, indeed, the observations 

 of the same day evidenced the nature of the change produced by the dis- 

 turbance; the times observed before the commencement of the perturbation 

 being sensibly longer than those which were determined during its con- 

 tinuance : thus the following times of vibration were observed on the 23rd 

 of November. 



Before the disturbance. After it had commenced. 



330 Time of vibr. = 12-1569 9 6 Time of vibr. = 12*0724 



4 30 = 12-1670 10 = 12 0894 



6 = 12-1531 11 = 12-0752 



8 = 12-1424 12 = 12-0624 



13 = 12-0901 



The following day the time of vibration returned to its previous dura- 

 tion. 



