66 



M A < ! N V. TIC S U It V E Y O V V E X N SYL V \ SI A. 



By comparing the differences in latitude and the corresponding differences in 

 dip for each place with the mean values of the group, their general accordance was 

 ascertained. None of the differences were large enough to require an exclusion 

 from the series. It need hardly be remarked that a slight consideration shows that 

 the dip depends almost exclusively upon the latitude ; the longitude factors will, 

 therefore, necessarily he very small. 



Method of Discussion. — The interpolation formula, proposed by the Rev. H. Lloyd 

 in 1838 (see the eighth report of the British Association, Vol. VII, p. 91), will be 

 used here in a slightly altered form, to allow for the convergence of the meridians. 



Let / = resulting dip or inclination. 



I Q = assumed mean dip for the epoch adopted (1842.0), and the mean lati- 

 tude and longitude, i its correction. 

 dL = difference of latitude, dM= difference of longitude, x, y, z,j>, q, as well 

 as i are to be determined by application of the method of least 

 squares, from the observations themselves. 

 /= I Q + i + xdL + ydMcos L + zdL dM cos L + pdD + qdJIP cos 1 L. 



Correction to epoch. — The mean epoch of the six groups is November,, 1841, for 

 which we can substitute without material loss of accuracy January, 1842 (or 1842.0). 

 Comparing the observations made by Mr. Schott, in July and August, 1862, with 

 the corresponding observations about the epoch 1842, we have the following table 

 of differences of residts for an interval of nearly 20 years: — 



