2 On the general distribution of Terrestrial 



The area under discussion is so large, and the observations 

 comparatively so sparse, that nothing more than the general dis- 

 tribution can at present be attempted. Local deviations from 

 the general system, of greater or less magnitude and extent, are 

 apparent in the table of residuals given at the close of this pa- 

 per; which m'ust be ascribed mainly to local attraction, since the 

 errors of observation are far less in amount, and point out local- 

 ities where additional observations will be most useful. 



The results of the Coast Survey observations are given in 

 Table I, which gives the latitvide and longitude of the stations, 

 the declination, dip, and horizontal intensity of the earth's mag- 

 netic force, the date of the observations, and a reference to the 

 particular locality, its geology, and other circumstances. 



Tlie record of these observations and the details of methods 

 and instrumental constants will shortly be published as a part of 

 the Coast Survey records and results, for the publication of which 

 Congress has provided. A brief notice will therefore suffice here. 



In observing the declination^ the magnetic meridian has gene- 

 rally been obtained by means of collimator magnets, using Gauss 

 and Weber's transportable magnetometer; while the astronomi- 

 cal meridian was derived from the triangle sides of the Coast 

 Survey, or obtained by direct observations. 



The dip has been observed with needles of from six to ten 

 inches in length, made by Gambey and by Barrow. Two nee- 

 dles have generally been used; or when one only was employed 

 it has been carefully tested and compared. 



The horizontal intensity has been determined in absolute meas- 

 ure by vibrations and deflections, according to the methods of 

 Gauss and Lamont The units of measure are those used in the 

 British surveys. 



From the agreement of repeated observations, it is inferred 

 that the uncertainty of the observations at a particular spot does 

 not exceed one or two minutes of arc in the declination and dip, 

 and j~J-^ part of the horizontal force. 



The data derived from other sources that are combined with 

 the Coast Survey observations are all of recent date, in order 

 not to introduce much uncertainty into the reduction to a com- 

 mon period. They are : 



1. Observations by Lieutenant (now Colonel) Lefroy, of the 

 Royal Artillery, in Canada, along the St. Lawrence, and at To- 

 ronto; being part of those published by Colonel Sabine, in the 



PkilosopJiical Transactions, 1846 and 1849, 



2. Observations made in connection with tlie survey of the 

 northeastern boundary ; ibid. 



3. Observations of horizontal intensity in Waterville, Maine, 

 by Professor G. W. Keeley, in 1847; Phil Trans, 1848, 



4. Observations by the late Dr. John Locke in various parts 

 of the United States, especially in Ohio and the northwestern 



^ 



