A Magnetic Survey of the Fortieth Parallel. 137 



his Magnetic Survey of Scotland in 1856-57 : it had been carefully 

 examined by Dr. Balfour Stewart for error of axle and evidence of 

 local magnetism in the circle. (See " British Association Beports," 

 Aberdeen, 1859.) After a number of trials with both circles, I 

 decided to make use of the one belonging to Owens College (the 

 Dover circle), principally for the reason that the method of lifting 

 the needles in that instrument seemed to afford more concordant 

 results than that adopted in the other. The Dover circle was pro- 

 vided with two needles of the ordinary form, each 3" 5 inches long and 

 0*27 inch in maximum breadth. 



In order to preserve them from rust during the sea-voyage, the 

 needles were carefully wrapped in gutta-percha tissue and placed in a 

 small box filled with soda-lime. This method of preservation was 

 found to be perfectly efficacious : at no time during the journey was 

 the slightest trace of oxidation discernible under the magnifying 

 glass. 



The mode of observing was that described by Mr. Welsh in the 

 " Admiralty Manual of Scientific Inquiry," 3rd edition, 1859. Eight 

 readings of the vertical needle gave the magnetic prime vertical. 

 The circle was then turned so that the needle could oscillate in the 

 magnetic meridian. Sets of readings were taken with the face of the 

 circle first east and then west, the face of the needle being towards 

 the face of the instrument; the needle was then reversed on its 

 bearings and again read east and west : afterwards its polarity was 

 reversed by stroking with the bar-magnets and both series of readings 

 were repeated. The needle was raised and lowered several times 

 between each set of readings ; these were in all cases repeated, and if 

 a difference of more than 3' was observed, a fresh set was taken. 



A comparison, made at Kew, between the Dover circle and the 

 instrument employed in the monthly observations regularly carried on 

 at that observatory, afforded the following results : — 



Kew, June 7, 1878. Time, 9.30 a.m. to 12.26 p.m. 



Needle 1. Needle 2. Mean. 



Dover Circle 67 41-8 . . 6°7 44*1 . . 67 43'0 



■ 67 43 ' 8 . . 67 44*1 



KewClr0le <" - 67 447 .. 67 44-1 . > 67 44 2 



f 1st series. . 

 [2nd „ .. 



A second set of comparisons at Kew, made after the return from 

 America, gave the following numbers : — 



Needle 1. Needle 2. Mean. 



Dover Circle, Oct. 2, 1878 67 457 .. 67 41*6 .. 67 437 



-**{TS: 5 SI SS ::} 6!1 « 



