Frazer.] 



050 



[April 6, 1877. 



amount expressing tl»e diflerence between the two declinations calculated 

 and noted for each term of years between such data. When observations 

 are entirely wanting, an approximation to the true declination may be 

 made in the manner already described; though the accuracy of such a me- 

 thod will be inversely proportional to the distance along the isogonic nor- 

 mals to old stations of record, and directly as the distance apart of adjacent 

 curves. 



a. Interpolation from Assistant Schotf s tables of decennial values of luagnetio 

 variation, U. S. Coast Survey Report for 1874. ■* 



b. From chart of Isogonic linos 1870. The positions of the curves with respect 

 to Mount Holly were measured from additions in pen by Mr. Schofl. 



c. Observation by Beaton Smith. 



