INTRODUCTION. XXVII 



The change of magnetic moment for one degree of temperature, or the value to be employed 

 as a constant coefficient, is calculated by Captain Riddell, according to the formula 



a n cot w. 



representing the angle of deflection at lowest mean temperature. 



a the arc value of one scale division in terms of radius. 



n the difference of scale readings corrected for change of declination. 



t t the change in the temperature of the deflecting magnet. 



Substituting the numerical values found by the preceding experiments, and adopting as the 

 total angle of deflection the differences of circle reading after the temperature observations, be- 

 cause most coincident in time, we have 



u = 5 48' 20" cot .... 0.992789 



a =.000415 log .... 6.618048 



n = 5'.987 log .... 0.777209 



t 1 = 62.06 log 1.792812 . . ar. co ... 8.207188 



1 log .... 0.000000 



q = 0.00039376 6.595234 



A portion of the earlier observations, which follow, were made by myself, as well as those at 

 Valparaiso, Herradura, Caldera, Copiapo, and La Candelaria mine ; all of the others by Lieut. 

 MacRae. 



At Valparaiso the observatory grounds of Mr. Mowatt, on the hill, above the custom-house, 

 were selected for the station ; at Herradura the grounds of the English Copper Smelting Com- 

 pany, near Captain Fitzroy's old site ; at Caldera the terrace in front of the custom-house ; at 

 Copiapo the patio to the house of Don Diego Carvallo, half a square S.W. of the plaza ; and 

 at La Candelaria, near the mouth of the mine. This mine is on the summit of a hill called El 

 Bolaco, 3,698 feet above the level of the ocean, in latitude 27 47' 51" S. ; longitude 4A. 41m. 

 57*. W. 



Occasionally, on the first day of the month, the unifilar magnetometer was set up, and the 

 deflector bar being placed perpendicular to and within a short distance of the suspended magnet, 

 changes of the latter were observed during an hour at intervals of 10*. These observations will 

 be found on pages 27 31. As the two magnets were accurately in the same horizontal plane, 

 and the unifilar was turned in the opposite direction through an angle precisely equal to the 

 deflection caused by the stationary magnet, this angle and the time of vibration of the latter 

 being known, fluctuations of the horizontal force may be calculated. The requisite data will be 

 found on pages 32 117. 



All the calculations on pages 120 123 were made by Lieutenant Archibald MacRae, princi- 

 pal assistant. 



THE INCLINOMETER. 



DESCRIPTION OF THE INSTRUMENT. 



The instrument used in determining the dip was made by Messrs. Henry Barrow & Co., 

 Oxendon street, London. It is a modification by Professor Kreil and Dr. Lloyd, of the circle 

 devised by Gauss. It consists of a four-inch horizontal circle divided to half degrees, read by 

 means of a vernier to minutes, and which rests on a centre-piece with three levelling screws. An 

 oblong brass plate revolving above the circle carries four upright columns, one pair of which, 

 three and a half inches high, are within a flat mahogany box, six and three-quarter inches high 



