Variation and Dip of the Magnetic Needle. 49 
In order to determine what changes the dip is undergoing, I 
first drew upon a chart lines representing the recent observations, 
and from these determined the dip at each of the stations where 
observations were made in Long’s Expedition, and which are 
given in my former paper. The mean difference is 30’. In the 
absence then of more accurate data, we may call this the dimi- 
nution of the dip from 1819 to 1839, being at the rate of 1/5 
per year. 
The observations at New York also indicate a diminution since 
1822, but being made with different instruments, and by differ- 
ent individuals, they are scarcely comparable. In order however 
to arrive at the most satisfactory result, let us take the mean of 
the observations by Capt. Sabine and Sir John Franklin, con- 
tained in my former paper, which gives us 73° 16’, cortestoriditiy 
to 1823.5. ‘Taking also the mean of the observations by Capt. 
Back, Prof. Bache and myself, we have 72° 51’ corresponding to 
1835.3. The difference is 25’ for 11.8 years, indicating a dimi- 
nution of 2./1 per year. The mean between this and the former 
determination is 1.8, which I assume to be the annual diminu- 
tion of dip for the United States. The observations contained 
in this and my former paper were now all reduced to the year 
1840, by applying the annual variation, and the lines of equal 
dip drawn upon the accompanying chart. Among the thirty ob- 
servations by myself, nine appear to be in error to the amount 
of at least 10’. They are as follows: 
Ann Arbor, error — 19’ Ypsilanti, error — 15 
Pittsburgh, +17 Monroe, +13. 
Z +16 IT ig ae S eee 
Baltimore, — +16 : Princeton, » mae See 
Albany, 15. 
The lines of equal dip were first deen upon a large and accu- 
rate map of the United States, and from this were copied upon 
the accompanying chart. The errors here mentioned were meas-= 
ured upon the original map, and not upon the accompanying chart, 
on which the position of several places is marked erroneously. 
The above differences are to be ascribed to local attraction, and 
errors of observation. Errors of this kind are unavoidable. The 
magnetic survey of Scotland, by Major Sabine, exhibits a greater 
number of errors of this magnitude, and the sum of the errors is 
also greater. 
Vol. xxxix, No. 1.—April-June, 1840. 7 
