TRANSACTIONS OF THE SECTIONS. 
2y 
ai whilst the needle was undisturbed in its case. The result was always the satne ; 
If'naenet always indicated increase of strength, which diminished to rts original 
csoto so soon as ever the light was withdrawn. This uniformity of result m all 
newquite incompatible with any supposition that the disturbance of the magnet 
FWCM^hycurr^ta. In order tu complete the theory of the vertical force, itwas 
t*»ary to obtain the coclBcient of reduction. The method of Dr. Lloyd was ex- 
Mdit BanibaT to peculiar objections, particularly on account of the little dqiend- 
aatkii could be placed oo the determination of the dip, which could not be safe 
oiasd^ee, but which enters as an important quantity into Dr.Lloyd s formula, 
imtalraetbodwasthemforeueed. The latter had dm advantage of allowing fre- 
ttpetition; and therefore, although not superior to Dr. Lloyd s m othw le- 
fcu. permitted much greater accuracy to be obtained in the dcternimation of the 
Bdcimt. This metboil wns invented only at the close of last year; hut since that 
b(idaily detemiiuatiun of the coefficient has been made. The effect of tempera- 
Sonpon all the magnets used in theobacrvalorv under ordinary atmoapheric changes 
aibeeo proved tn be insensible. It had further been shown that the diminution of 
ckaetism by increase of temperature was not due to the quantify of heat imparted, 
m the velocity with which it was imparted. After the heat is withdrawn, the 
“Boation still contuiuta; and after a time the magnetism ^adually retunm into 
iaagDet as before the application of heat. These discoveries explained why the 
»Wion for temi>erature under ordinary circumsinncea had provwl to be nothing. 
IVe Axrmation at Bombav is only about 11 ' Knat. The scale of the decluiome^r is 
ttsAre read by a transit instrument ; and tho readings had never gone out of t^he 
ittofthe telescope even in the largest distarboncea. Thus each observation be- 
<m an absolute determination. The declination was generaUy inovmg eastward 
^fhout 1845. The diurnal changes of declination aie not of ibe same character 
^ghout the year. From the Ist of April to the 8th of October, the most easterly 
f-banofthe magnet's north cud is at l6'‘ G.M.T. It then moves through about 
tsyvstwarduntil iob. Next back again through about 2'‘a until 0 orl • Auen 
ttpin moves westward until 3*'. After this time it slightly rises ; but from 5 to 
[Bombay night) it remains nearly in its mean position j and then m me course 
■^J^moveg through half its win i ■ range to its extreme easterly positiou. But about 
'Printer solstice the curve h8.‘ .suroed n different finite. It has now o^o extreine 
^l/pnfiitioo at O**, and tme extreme westerly position at From the early 
Wof October to the early part of December, the changes gradually pass from the 
•■tothf second of iho&e above described. , 
UK burizoDtal force has tlic same character of diurnal change throughout the 
^ d a mimmum about O'*, and a uiaKimum not far from the 17 th hour. e 
l^tal force, which had been determined every day nearly by the Gaussian, me- 
^WBsiliffliniBbing. The horizontal force magoptometer gave the same result. 
5 two however did not agree in details, and the Gaussian method seemed least 
The Gaussittn calculations however proved that the magnetisms ot all 
J* Wdlw wi-re gradually dimiuislung, and oUier experimeuts seemed to show ttiat 
■'tonasgitpoutmagnatUni, some quickly and some slowly. 
^ did not allow more than s reference to some cf the more iropor^t steps 
had be«m taken in niclcurcilogical inquiry. The actinometer 
but the observations of 1841. 1842, 1845. 1846 , which 1 ^ been 
determine the doily curve, were not found to supply, even in the favo^uie 
2^here of Bombay, a sufficient number of observations m perfcctiy cl^ sky to 
7*any certain determination. The curve however appeals to be im hy perb 
J^nimeincal on each aide of an axis which represents the 
paiallcl to the axis of the htTerbola representing the heat, and 
^ Oae to the tangent of the vertex of the hyperbola 
The reduction of the temperature, in order to dclerrome lU laws, h^ bew 
« continuallv receiving heat from the 
Jjy cooling by radiation. Supposing these two cause* only to be m op^tion, 
S,^,^"'®‘'“ly-teccivt-d luwn of sol^ and terrestrial radiabon to 
a S « the hour of the day and the amount of ^ ^ 
surface may be formed and integrated without difficulty. Thus it may be 
