lxviili INTRODUCTION TO THE MAKERSTOUN OBSERVATIONS, 1845 anp 1846. 
124. From these we find for 
Professor BACHn’s bar, named | 63°-9 to 90°3 g=0:000095 { ae 8 ny 
B, marked 4, -+++.+-+e-+-+ } SB*1 to 63°9 ¢==0-000089 aman OF @ for 28, Babe 4 
Professor BACHE’S bar, named ) 657-4 to 89°°8 ¢g=0-000096 { =x 
jad unmarked, dete eds * 39°38 to 65°-4 q=0:000084 welecee occ e ates eeccetecesssscece 8 
Sir THOMAS BRISBANE’S ul 64°-7 to 89°°6 g=0:000106 { Seren _@ 
named M, unmarked, ------ 39°8 to 64°°7 g=0-000096 10 
Professor FoRBES’s bar, named | 64°:8 to 88°:2 g=0:000295 { ag 
F, unmarked, ..:..........-- 39°:5:/10164-Sig = O.000816 snares ait fa 14 
Sir THoMAS BRISBANE’S bar, | 67°:8 to 94°1 g =0-:000200 { sing 
marked I. 23, ......... onze (PGOro tolGde- Sg OOU0 Meta) po "ecco sae ae 13 
125. For these magnets, therefore, the temperature coefficient varies from + 
io J, of its mean value from the mean temperature of 50° to that of 75°, the co- 
efficient being greater at high temperatures than at low temperatures. A similar 
result was obtained by Mr CurisTI£ long ago,* and lately by Mr Arry and Colonel 
SasinE. This difference has not been found to exist for the large variation magnets 
(see Introduction, p. xliv.). 
126. The following are the results of series of observations for the temperature 
coefficients of different magnets. 
TABLE 15.—Results of Observations for the Temperature Coefficients of different 
Magnets, the Temperature Rising and Falling, made 1843-7. 
Temperature ° 
P Difference 
Description of Magnet. UAE oes Bisg 
minus 
q: q Falling. 
Noy. 9, 1843} Bifilar Magnet (15-inch) ------++.+--+++ Beds ceies 0-000290 | 0-000298 | + 0.000008 
Deflection Magnet (15-inch) 0-000278 | 0-000296 |+ 0.000018 
Balance Magnet (12-inch, very thin) 0-000067 | 0-000079 | + 0-000012 
3-65-inch Magnet, named F. 
3-65-inch Magnet, unmarked 
0-000280 | 0-000289 | + 0.000009 
0-000311 | 0-000301 | —0-000010 
0:000408 | 0-000395 | — 0.000013 
0-000747 | 0-000745 | — 0.000002 
0-000264 | 0-000278 | +0-000014 
0-000210 | 0-000212|+0-000002 
0-000323 | 0-000333 |+0-000010 
0-000100 | 0-000121 |+0-000021 
0-000091 | 0:000091} 0-000000 
0-000089 | 0.000092 | + 0-000003 
0:000098 | 0-000104 | + 0-000006 
3-inch Magnet, marked 8, 43. -+---.+-++- 
3-inch Magnet, marked S. 29 «-+.+++.++eeeeeee+ 
3-inch Magnet, hollow, marked B, 6, -+++++++- 
5-inch Magnet, hollow, marked I, 23, ---+-++- 
st 3-65-inch Magnet, named F. oo 
May 28, 1847 | 3-65-inch Collimator, named M. «--.-++---++++: 
Sept. 8, 1847) 3-65 Collimator Magnet, marked 4, named B. 
3-65 Collimator Magnet, unmarked, named B.— 
3-65 Collimator Magnet, named M. 
WWRANAAAANSD 
The observations in some instances are too few, and with too large probable 
errors to be sufficient for the determination of the difference of the coefficient, for 
rature in handling during the removal of the bar, an error would be introduced always of the same 
sign if the north end has always the same position at the beginning. 
* Philosophical Transactions, 1825, p. 63. 
