50 



REPORT ON 



The constants, peculiar to the portable declinometer D 22, were obtained as 

 follows : 



The Temperature Coefficients of the magnets were furnished by Mr. Chas. A. 

 Schott, of the U. S. Coast Survey. They had been used with the instrument 

 for some years, and I had no opportunity to redetermine them. They are as 



follows : 



For the magnet C 32 q = 0.00020 



" " " S 8 q = 0.00027 



In reducing the observations a correction was always applied to the magnetic 

 moment of the magnet C 32 to reduce it to what it would have been if C 32 had had 

 the same temperature as S 8. Hence, the temperature coefficient of C 32 was the 

 only one used, and in order to facilitate its application the following table was com- 

 puted which furnishes the value of log. [1 — (t' — t) q] with the argument (t'^t). 



Correction of Magnet C. 32 for Temperature 



if'-t) 



Log. [1 — {t>— t)q\ 



{t'-t) 



Log. [I— [t'—t) q\ 



4- 1° 



9.99991 



— i° 



O.OOOO9 



+ 2 



9.99983 



— 2 



O.OOO17 



+ 3 



9-99974 



— 3 



O.OOO26 



+ 4 



9.99965 



— 4- 



O.OOO35 



+ 5 



9-99957 



— 5 



O.OO043 



+ 6 



9.99948 



— 6 



O.OO052 



+ 7 



9-99939 



— 7 



O.OO061 



+ 8 



9.99930 



— -8 



O.OO069 



+ 9 



9.99922 



— 9 



0.00078 



+ xo 



9-999I3 



— 10 



0.00087 



O. I 



1 



0.2 



2 



°-3 



3 



0.4 



4 



°-5 



4 



0.6 



5 



0.7 



6 



0.8 



7 



0.9 



8 



The Value of One Division of the Magnet Scale was determined for each magnet 

 in the following manner: The instruments having been set up and adjusted as 

 usual, the magnet was suspended in the magnet-box, and the packing blocks 

 (before described as being used to prevent the suspension fibres from being twisted 

 when the instrument was packed for travelling) were inserted in such a manner as 

 to hold it perfectly steady. Then, the magnet scale being horizontal, the vertical 

 wire of the theodolite telescope was made to coincide with any convenient scale 

 division, and the horizontal circle of the theodolite was read. Next, the vertical 

 wire was made to coincide with some other scale division, and the circle was again 

 read. The difference of the two circle readings, divided by the difference of the 

 two scale readings, gave the angular value of one scale division. 



The following are the observations in detail for each magnet : 



