Magnetism in the United States. • 7 



that hitherto there have been no observed facts, in the region of 

 Texas and New Mexico at least, upon which a decision could be 

 based, 



A form of branching similar to that on Barlow^s map, and at 

 variance with Gauss' theory, occurs in Hansteeris Erdmagnetis' 

 musj and has been preserved on a recent map of the declination 

 in his Magnetishe Inclinations Forandring^ 1855. 



It will be seen that if Barlow^s system represents the forms 

 correctly, the line of 9"^ easterly declination along the coast of 

 Texas should turn to the southeast; while the line on our map, 

 fixed by numerous observations, decidedly turns to the southivest^ 

 in conformity with Gauss's system. In going westward along 

 the Mexican boundary, there is not a decrease and subsequent 

 increase, as Barlow's map indicates, but a steady increase, estab- 

 lished by the observations along that boundary. The general 

 agreement in form between the lines on our map and the Gauss 

 system is striking. Observations in Central America would at 

 present be of the greatest value in deciding the matter imder 

 discussion. The isolated observation at Panama, Table III, has 

 assisted greatly in fixing the direction of the line of 1^ east, de- 

 termining that it still turns to the southeast. 



On the Atlantic coast the Gauss lines of declination agree well 

 in position with ours, allowing for the difference of epoch. On 

 the Western Coast, they are farther to the south, which seems 

 contrary to what is supposed to be the secular change. 



Dip, — With Loomis map, before quoted, and Sabine's map of 

 the dip in British America {Phil, Trans,, 1846), the agreement is 

 excellent. The Gauss lines agree Avell in form, but differ from 

 2° to 2^° in the amount of dip. This is due to the errors of 

 Horner's map, from w^hich they are derived, and on which the 

 lines are about 2^^ out of position. It must be remembered 

 that they were altogether interpolated, no observations of dip in 

 the United States being known when Ilorner's map was con- 

 structed. 



Horizontal Intensity, — For a comparison of this element we 

 have only Gauss's chart, on which the lines are affected by the 

 erroi^ of the dip lines, by means of which they have been de- 

 rived from Sabine's general map of the total intensity for 1836. 

 The values of the Gauss lines being ex}>ressed in the arbitrary 

 scale, multiplied by 1,000, we reduce them to absolute measure 

 through the observations at Toronto, where, in 1844, the total 

 intensity in the arbitrary scale was 1-836, the dip 75"" 13''4, and 

 the horizontal force S'54, in absolute measure in British units, 

 giving a multiplier of 7*5G for the arbitrary scale, or 0*00756 for 

 Gauss values. We find thus 3'78 equivalent to Gausses line of 

 500, along whieli tlie actual horizontal force is 4-24 to 4*30 ; in 



the same place the dip is really 71° 30', and on Gauss' map, 73'' 



1 



-1 



■i 



15'. Corresponding differences exist in other parts of the map. 



/ 



I 



