-M A ( i \ E T I C S U R V E Y O P J' E N N S Y L V A N I A . 



67 



The increase in the dip is therefore very slight, and if we consider that, accord- 

 ing to Mr. Schott's investigation (Appendix, No. 32, Coasl Survey Beporl for L856) 



the dip near the Atlantic coast, about the years 1841—184 I. was at its minimum 

 value, and hence could not have changed sensibly fur several years, we can without 

 any sacrifice of accuracy in our reduction, use our results as if all belonging to the 

 mean epoch 1842.0. No reduction to epoch has therefore been applied. It is pro- 

 bable that the present annual increase amounts to about 1'. At Toronto, between 

 1844 and 1855 (sec Vol. Ill), the annual increase was 0'.8. 



In the formula of interpolation, 1 retain the factor cos L, thus making it com- 

 parable with similar expressions, for other localities, where the introduction of cos 

 L may be more important.' 



The value of the magnetic survey of Pennsylvania is increased from the fact that 

 the isoclinal lines are presented for an epoch at which the dip was probably near 

 its minimum value. 



The conditional equations are of the form — 



=I Q —I+ i + xdL + ydMcos L + zdL dMcos L + pdL 2 + qdM 2 cos 2 L. 

 We find from the solution of the normal equations, the expression, 

 i = 73°.26 + 0.876 c/Z-0.076 dMcos Z-0.023 dL dMcos L + 0.007 dL 2 + 0.013 



dJP cos 2 L 

 where dL = lat. — 4P.38 

 dM= lone.— 77.58. 



The preceding- investigation was made for the purpose of ascertaining what terms 

 should be finally admitted in the discussion. 



Next, nine groups of five or six observations in each, arranged, in regard to their 

 geographical position and area, with as much regularity as the nature of the case 

 admits of, give, 



