Ifi 



\i a <; \ i: tic s D it v e y o P r e n n s y l v a n i a. 



The small extent of the survey, as well as the comparatively small number of 

 observations, will not permit the introduction of curvature in the isogonic lines; 

 they are therefore treated as straight lines. This assumption also serves for the 

 recognition of any local disturbances as indicated by the differences of obsemd 

 and computed values. 



Let D = + 2°.08 + x dL + y dMcos L 

 Where dL = Lat. -40°.98 



dM = Long.- 77.95 



The 16 conditional equations have been formed, and the values of x y & D found 

 from the normal equations, are as follows: — 

 x -- + 0.5 10-2 

 y - - 1.206 

 D= + 2°.08 + 0.5102 dL— 1.206 dMcos L. 

 A comparison of the observed and computed declinations shows the necessity of 

 introducing a term involving dL dMcos L; this lias been done and the solution of 

 the normal ('([nations "•ives us the following expression: — 



D = + 2°.14 + 0.513 dL — 1.231 dMcos L — 0.203 dL dMcos L. 



