140 



EIGHTH REPORT — 1838. 



bury and London (near the commencement and end of the se- 

 ries, respectively), as compared with the difference observed at 

 the same places in spring. 



With respect to Needle L. 4, the observations at Dublin in 

 April and May show that its magnetism was perfectly steady 

 during the spring series. This needle, however, sustained a 

 very great loss of force between the two sets of observations 

 with it at Ryde ; and this loss appears to have been, in a great 

 measure, a sudden one. But that the magnetism of the needle 

 was not stationary during the remainder of the autumn series, 

 will appear at once from a comparison of the observations at 

 Shrewsbury in April and August. As we have no satisfactory 

 means of determining the amount of this loss, and of interpo- 

 lating a correction, we are forced to reject all the results ob- 

 tained with this needle in autumn. 



Table XLIV. 

 Computed Intensity. 



Means. 



Shrewsbury 1-0077 



Holyhead 10144 



Ryde 0-9972 



We have here eighteen results at the twelve stations, which 

 being combined by the method of least squares, give the following 

 values : a: = + -000047 ; y = — '000067 ; ?< = - 54° 49' ; r = 

 •000082 ; the mean geographical position being lat. 52° 0', and 

 long. 1° 50' W., at which the probable value of the intensity is 

 1-0048. 



