Contributions to Terrestrial Magnetism.-No .
VI .
By Lieut.-Colonel EDWARD SABINE , R.A. , F.R.S. Received April 17 , -Read April 18 , 1844 .
Q 10 .
Observations made on Board Her Majesty 's Ships Erebus and Terror , from June 1841 to August 1842 , in the Antarctic Expedition under the command of Captain Sir JAMES CLARK Ross , R.N. , F.R.S. I HAVE now to lay before the Royal Society the results of the Magnetic Observations made at sea by the Antarctic Expedition during the second year of its operations in the southern hemisphere .
Leaving Hobarton early in July 1841 , the ships proceeded in the first instance to Sydney in Australia , and from thence to the Bay of Islands in New Zealand , where they remained until the return of the season of navigation in the high latitudes .
Quitting New Zealand in November , the ice was met with and entered in a somiewhlat lower latitude than in the preceding year , and in a longitude considerably to the east of the former track .
The obstacles which the ice presented to their progress appear to have been greater than on the former occasion ; they were however surmounted , and in February 1842 the ships again reached the ice barrier , or glacier , in latitude 78 ?
, by which they had been stopped in the preceding year .
After an unsuccessful endeavour to turn the eastern extremity of the glacier , the advance of the season compelled their return to the lower latitudes ; they quitted the Antarctic Circle in March 1842 , and keeping nearly in the 60th parallel , crossed the whole breadth of the southern Pacific Ocean to the Falkland Islands , where they arrived in April .
I proceed at once to the examination in detail of the magnetic observations made during this period .
Deductions of the Constants a and b in the Correctionsfor the Ship 's attraction .
1 .
In the Erebus.-For the constants a and b to be employed in computing the corrections of the declination , we have the observations on each of the 32 principal points of the compass at Hobarton , in October 1840 and June 1841 .
We have also a similar series at Port Louis , in the Falkland Islands , in August 1842 .
The observations at Hobarton have been already discussed in No. V.* Those at Port Louis were as follows : August 19 , 1842 .
Ship 's head Disturbance Ships head Disturbance Ship 's head Disturbance Ship 's head Disturbance topiartowadds th t wiardsthe towiards the by compass .
towards the by copass towards the by copass .
towards the by compass .
towards the __ west .
west .
west .
west .
es N. +0 127 w. -2 15.8 s. +0 00'1 E. +2 074 N. by w. -0 04'1 w. by s. -2 21.2 s. by E. +0 43'9 E. by N. +1 54'0 N.N.W. -0 33'6 w.s.w. -2 21'3 s.s.E. +1 1 '7 E.N.E. +1 44'0 N.W. by N. -0 50'1 s.w. by w. -2 4*3 S.E. by s. +1 41'4 t N.bE .
E. +1 16'5 N.w. -1 02'3 s.w. --1 80 s.E. -+1 55 5 N.E. +0 50'9 N.w.by w. --1 00*6 s.w.by s. -1 3'3 S.E. by E. +2 06'9 N.E. by N. +0 40*5 W.N.W. M -1 49-3 s.s.w. -1 17'3 E.SN .
E. + 189 !
E +0 41'2 w. by N. -2 09'6 s. by w. -0 3856 E. by s +2 164 , i N. by E. +0 277 The values of the constants deduced from the observations at Hobarton were , a=+'0272 ; b6=+'986 .
The values fioml the observations at the Falkland Islands are , a=+-'0292 ; b= +'984 .
The values of a at Hobarton were derived from t'wo series , one in October 1840 , when the ship had recently passed through the low magnetic latitudes , and the other in June 1841 , on her return from the highest magnetic latitudes of the southern hemisphere ; the two series separately considered give a=+'0235 in 1840 , and #0309 in 1841 ; we have therefore the following values+-0267 in the Thames , where the ship had been stationary for several years .
+'0235 at Hobarton , on her first arrival from the low latitudes .
+ 0305 on her return to Hobarton from the very high southern magnetic latitudes .
+'0292 at the Falkland Islands in 1842 , on her second return from the very high southern latitudes .
The variations in these values is in accordance with the view expressed in the preceding Number of these Contributionst , that when a ship changes her magnetic latitude , the corresponding change in the induced portion of her magnetism may not be instantaneous ; that some portions of her iron may be of a quality intermediate between perfectly soft iron , which would undergo instantaneous change , and iron permlanently magnetic ; and that when changing rapidly her geographical position , she may be liable to be more or less in arrear , in regard to her magnetic condition , of her actual locality at any particular time .
In a ship in which this should be the case , a table computed with any one value of a would not apply equally to one portion * Philosophical Transactions , 1843 , Part II .
pp. 152-154 .
88 t Ibid. pp. 152 , 153 .
of her voyage in which she might be sailing from lower into higher inclinations , and to another portion in which she might be returning from higher into lower magnetic latitudes .
The voyage under consideration comprised two such portions ; and I have therefore employed two tables for the Erebus , one computed with '0267 for the period when the ship was increasing the dip , and the other with '0288 for the period when she was decreasing the dip .
The differences are insignificant , except when the inclination is very high ; the greater part of the declinations observed in the high dips were antecedent to the 1st of March 1842 , when the ship commenced her return to the lower latitudes ; for these the table computed with a='0267 has been employed , and appears to answer better than the corrections computed either by the values resulting from the observations at Hobarton before the commencement , or by those at the Falkland Islands after the conclusion of the voyage .
2 .
In the lerror.-For the values of a and b in the Terror , we have observations on each of the thirty-two principal points of the compass at HIobarton in October 1840 , and a second series in June 1841 , as follows : Ship 's head by Disturbance towards the west .
Ship 's head by Disturbance towards the west .
compass .
compass .
1840 .
1841 .
Mean .
1840 .
1841 .
Mean .
N. +041.4 0 .
/ 05 10 / 0I N. +0 42-4 -0 52 -0 05 s. -0 11.6 -0 55 -0 33 N.byw .
--0 23-6 -0 52 -0 38 s. byE .
+0 524 -0 06 +0 23 N.N.W. -1 20'6 -0 59 -1 10 s.s.E. +1 56'4 +0 43 +1 20 N.w. by N. '-2 206 -0 03 -1 12 s.E. by s. +2 38-4 +2 08 +2 23 N.W. -3 25-6 -0 58 -2 12 S.E. +3 19-4 +2 57 +3 08 N.W. by w. -3 56*6 -2 12 -3 04 S.E. by E. +4 00'4 +3 48 +3 54 W.N.W. -4 01-6 -2 26 -3 14 E.S.E. +4 43-4 +5 25 +4 54 w.by N. -4 06-6 -251 -3 29 E. by s. +4 28-4 +4 58 +4 43 w. -4 36-6 -3 34 -4 06 E. +4 24-4 +4 27 +4 26 w. by s. -4 44-6 -3 43 -4 14 E. by N. +4 4 11 4 02 +4 07 w.s.w. -4 52.6 -4 34 -4 43 E.N.E. +4 07-4 +3 27 +3 47 s.w.byw .
-5 22*6 -4 01 -4 42 N.E. by E. +3 27'4 +3 04 +3 16 s.w. -4 23-6 -3 50 -4 07 N.E. +3 02*4 +3 01 +3 02 s.w. by s. -3 31-6 -4 22 -3 57 N.E. by N. +2 37'4 +2 27 +2 32 s.s.w. -2 03*6 -3 41 -2 52 N.N.E. +2 11*4 +0 46 +1 29 s.byw .
-1 37-6 -2 44 -1 11 N. by E. +1 26-4 -0 12 +0 37 We have also a series at Port Louis , in the Falkland Islands , in August 1842 , as follows : -Ship 's head by Disturbance lShip 's head by Disturbance Ship 's head by Disturbance Ships head by Dsturbance towards the towards the towards th towards the compass , tow he compasst.to t west.he coasmpass .
twst -N .
r0 19 w0 / -0 E. 6 N. +0 19 w. -2 30 S. -0 16 .
+2 46 N. by w. -0 02 w. by s. -2 21 s. by E. -0 08 E. by N. +2 27 N.N.W. -0 17 ws.w .
-2 1W2 S.S.E. 0 00 E.N.E. +1 58 N.w. bin .
-0 48 s.w. byw .
-2 21 s.E. by E. +0 47 N.E. by E. +1 39 .
N.W. -1 19 s.w. -1 33 S.E. +1 35 N.E. +1 13 N.w. by w. -1 49 s.w. by s. -1 05 s.E. by E. +2 17 N.E. by N. +1 11 W.N.W. -1 47 s.s.w. -0 47 E.S.E. +3 04 N.N.E. +0 34 w. by N. -2 07 s. by w. -0 45 E. by s. +2 33 N. by E. +0 27 N2 From these observations we have the following values of the constants : Hobarton ... ... a=+-0275 ; b=--'979 Falkland Islands ... a-+'0293 ; b=+-994 .
These values are nearly the same as those derived from the observations in the Erebus at the same periods , and appear to require no special remark ; the same tables have been employed in the declination corrections of both ships during the voyage under notice ; the values of the constants in these tables were as follows:a=*0267 when the ships were sailing from the lower into the higher latitudes ; a=-0288 when sailing from the higher into the lower latitudes ; b==+984 in both cases .
Deduction of the Corrections on account of the Ship 's attraction for the Observations of Inclination .
1 .
In the Erebus.-The spot in the ship in which Mr. Fox 's apparatus for the observations of inclination and intensity was employed , was a few feet in advance ( towards the bow ) , and about two feet lower in height , than the position of the standard compass .
The values of a and b derived from the observations with the compass needle apply in strictness only to the spot in which that compass was stationed ; it may be proper , therefore , before we employ them for the observations with Mr. Fox 's apparatus , to show that nearly similar values for the constant a in particular ( the more important constant ) are deducible from the observations of inclination and intensity , independently of those made with the compass needle .
For this purpose we may employ equation ( 1 .
) , Phil. Trans. , 1843 , Part II .
p. 147 , viz. cos 'c os c= os =c os cos +a sin , obtaining by its means the value of a from the observations of inclination and intensity made at Hobarton and Port Louis .
As A !
is known to differ very slightly , if at all , from unity , we have from equation ( 1 .
) , a sin 0= cos 4 ' cos '-cos 0 cos. p and 4 are furnished by the mean of the observations of inclination and intensity on the sixteen points of the compass , having approximate corrections applied to each of them ; p ' and 4 ' by the ( uncorrected ) observations on the different points .
Fromn the general aspect of the observations at both stations , we may conclude that the same symmetrical distribution of the iron existed in reference to the position of Mr. Fox 's apparatus as in the case of the standard compass , and consequently that at the north and south points the value of ' and C coincided , being equal in the one case to 0 , and in the other to 180 ? .
At Hobarton ( in June 1841 ) we have p= 1-83 , 4= -700 39 ' ; v ' at north 181 2 , at south 1-854 ; 8 ' at north 71 ? ?
56 ' , at south -69 ? ?
14 ' : 90 The correspondence in the value of the constants obtained from the observations at Hobarton and Port Louis , being the commencing and concluding stations of the voyage now under consideration , is fully as good as could be desired ; and a table formed from them has been employed for the correction of the observations made between Hobarton and the Bay of Islands , and during the return of the Expedition from the high latitudes to the Falkland Islands commencing with the 1st of March 1842 .
In thos- .
portions of the voyage the ship was passing from the higher to the lower magnetic latitudes , in which circumstance they corresponded with the observations at Hobarton and Port Louis , which were both made on the return from the vicinity of the magnetic pole .
But if we attempt to apply the same table to the observations made under the reverse circumstances , namely , when the ship was passing from the lower to the higher latitudes ( and such was the , case with the greater part of the observations which we have to correct in the present voyage ) , we find that the tabular numbers , where the N. and S. points are approached , furnish a decided over compensation .
On days when observations have been made at or near the N. and S. points , if we seek in the table for the corrections which should bring the results in accord with each other , we find that the corrections which will do so belong to a dip which is always some degrees less than the true terrestrial dip .
It appeared desirable , therefore , if possible,.to form a table for the correction of the observations of this portion of the voyage , derived from those observations themselves .
Fortunately we have a better opportunity of doing this than might have been anticipated .
The progress of the Expedition was so much impeded by ice in the early part of January 1842 , that from the 6th to the 16th inclusive , the Erebus was the whole time between the latitudes of 65 ? ?
54 ' and --66 ? ?
14 ' , and between the longitudes of 204 ? ?
33 ' and 202 ?
02 ' ; the weather and all other circumstances being favourable , the inclination was observed in the course of those eleven days with the ship 's head on seventeen different points of the compass , sufficiently distributed , and particularly towards the north points and south points , where the effect of the ship 's attraction is greatest , and is in opposite directions .
From the observations at north and south it is not difficult to obtain ant approximate value of a , which should bring the corrected resutlts at those points into accord .
The value thus obtained is about +'023 .
I have collected the observations during the period referred to into the following table , taking , for the sake of simplicity , only those observations which were made by the direct method , which , however , comprises by far the greater part of the observations of that period .
I have then computed the corrections , first , with the values of the constants , such as they are given by the observations made for their determination at Hobarton and the Falkland Islands ( being the commencement and close of the voyage ) , viz. a= +*028 ; b=+'984 ; c_+-015 and d=1 ; and second , with a-+'023 , b , c and d , as before ; and have placed the two series of corrected results in the table , with columns showing in both cases the difference of the corrected result , on each point , from the mean result .
A comparison of those columns seems conclusive in favour of the application 93 of the smaller value of a to those observations which were made when the saip was in progress from the lower to the higher latitudes .
If a be taken as it was found at Ilobarton and the Falkland Islands , not only are the differences generally greater , but they are systematically so ; evidencing an over compensation where the north and south points are approached ; whilst with the smaller value of a the differences are greatly diminished in amount , and exhibit no appearance whatsoever of system .
They are such as may well be supposed to have been occasioned partly Jby observation error , and partly by small differences of geographical position in which the observations thernselves were made .
Values of the Constants .
Values of the Constants .
a= +-028 .
a= + 023 .
Ship'shead Number Inclination b=+-984 ; c=+'015 ; d=1 .
b=+'984 ; c=+-015 ; d-1 .
bcp a ofobserby compass . ?
o. sel observed .
'I_ vabtons .
Computed Inclinations rComputed Inclinations corrections .
corrected .
corrections .
corrected .
` N. 1 -80 58 +1 32 -79 26 -20 +1 16 -79 4 -3 N.N.E. 2 -81 00 +1 27 -79 33 -13 +1 12 -79 48 +3 N.E. 2 -80 42 +1 12 -79 30 -26 +1 00 -79 42 3 N.W. 3 -80 35 +1 12 -79 23 -23 +1 00 -79 35 -10 N.E. by E. 2 -80 50 +1 01 -79 49 +3 +0 55 -79 55 +10 w. 1 -79 58 +0 17 -79 41 5 +0 14 -79 44 1 E. 3 -79 50 +0 17 -79 33 -13 +0 14 -79 36 9 E.bys .
1 -79 45 -0 01 -79 46 -00 -0 01 -79 46 +1 s.w.byw .
3 -79 19 -0 38 -79 57 +11 -0 31 -79 50 +5 s.w. w. 1 -79 30 -0 42 -80 12 +26 -0 34 -80 04 +19 s.w.-w .
1 -79 10 -0 46 -79 56 +10 -0 38 -79 48 +3 S.E. 1 -79 08 -0 55 -80 03 +-17 -0 45 -79 53 +8 S.W. 3 -78 52 -0 55 -79 47 +1 -0 45 --79 37 8 s.w.I- .
1 -78 48 -1 02 -79 50 +4 -0 50 -79 38 -7 S.S.E. 3 -78 28 -1 13 -79 41 5 -1 05 -79 33 -12 s. byw .
3 -78 28 -1 29 -79 57 +11 -1 13 -79 41 s. 5 -78 32 -1 31 -80 03 +17 -1 14 -79 46 +1 M|.eans 36 ... ... .
-79 465=a The mean of the observations in the table thus corrected is -79 ? ?
45 ' ; the corresponding geographical position is -66 ?
04 ' , and 203 ? ?
17"5 , if we take as such the middle point of the geographical space in which the ship was detained from the 6th to the 16th of January .
The inclination observed on the ice on the 16th of January , in lat. -65 ? ?
49 ' , long .
202 ?
02 ' , with needles whose poles were reversed , was -.790 39'-5 .
We can derive no precise conclusion in regard to the value of d , from observations which are not identical in locality ; but the accordance of the results obtained on board and on the ice , in geographical positions so little different , is quite sufficient to show that the error involved by assuming d as unity must be , at the utmost , very inconsiderable .
The tables for the correction of the inclination in the Erebus have therefore been computed with the following values for the constants , viz. froimi New Zealand to the end of February 1842 .
being ) the portion of the voyage in which the ship was in progress from the lower into the higher inclinations , a= +*023 , b-=+984 , c=+ 015 and d=1 : and for the remainder of the voyage a= +'028 , b , c and d , as before .
In the Terror.-The place in which Mr. Fox 's apparatus was used in the Terror was about the same distance from the position of the standard compass , and in the same direction , as in the Erebus .
A series of observations were made with it for the purpose of furnishing materials for the deterlnination of the constants , at Hobarton in June 1841 , and at the Falkland Islands in August 1842 ; and the inclination was also observed with the ship 's head on several points of the compass during the detention of the ships by the ice between the 6th and 16th of January 1842 .
In the case of the Erebus , we have found these latter observations of principal use in furnishing the values of the constants which apply to the greater part of the observations of the voyage ; it may , therefore , be advisable to commence with the discussion of the corresponding series in the Terror .
Inclinations observed on board Her Majesty 's ship Terror with needle F.C.B. used direct , during her detention by the ice from the 6th to the 16th of January 1842 , between the latitudes of -65 ? ?
45 ' and 660 20 ' , and longitudes of 2010 46 ' and 204 ?
04 ' .
Ship 's head Number of Inclination Ship 's head Number of Inclination by compass .
observations .
observed .
by compass .
observations .
observed .
N4 -81 19.5 s. 6 -78 30 N. E. 2 -81 14 s. W w. 1 -78 21 N. E. 1 -80 50 s. by w. 1 -78 48 N.N.E. 3 -80 57 s.w.byw .
Iw .
1 -78 50 N.E. 2 -80 48 s.w. by s. 3 -79 00 N.E. by E. 1 -80 26 s.w. 3 -79 08 E. ?
N. 1 -79 57 s.w. 2 w. 1 -79 08 E. 6 -79 55 s.w. by w. 5 -79 21 E. S. 1 -79 45 w.s.w. 2 -79 37 E. by s. 1 -79 33 w. by s. 1 -80 05 E.S.E. 2 -79 21 w. 1 s. 2 -80 07 S.E.byE .
E. 1 -79 04 N.W. 2 -81 09 S.S.E. 1 -78 42 N. by w. 1 -81 15 s. by E. 4 -78 37 These observations manifest the general systematic character of the disturbance occasioned by the ship 's attraction ; they furnish indeed a remarkable example of the success with which the effect of the ship 's iron on the inclination may be investigated by observations made at sea .
The disturbance appears to have not been strictly symmetrical , inasmuch as the inclinations observed on the western points somewhat exceed in amount those observed on the corresponding eastern points ; the same circumstance took place in the observations at Hobarton ; but at the Falkland Islands , on the contrary , the inclinations observed on the eastern points were generally somewhat the higher .
A similar occasional departure from strict symmetry has before been noticed in the effect of the ship 's iron on the compass needle* ; in that case also a Philosophical Transactions , 1843 , Part II .
p. 152 .
UX In IA " , A T. U 95 MJVlU A ; 2lI V. 0 the disturbance in the same ship was sometimes greater on the eastern , and sometimes on the western points ; these small irregularities , having no uniform character , are regarded as included amongst those varying accidents which are classed generally under the name of observation error .
It is proper , however , in consequence of this occasional irregularity , that the data from which constants are to be derived for general corrections should consist of the mean of observations on corresponding points on the east and west sides of the compass ; in this view we have as available observations in the preceding table those on the following points of the compass .
North ... ... . .
-81 195 N.W ... ... ... N.E. E.-.80 58'5 E. S. ( fro E. and E. S. ) W.byS ... ... ... E.by S ... ... .
W.S.W. .
-79 29 E.S.E ... .
E.S.E ... ... ** find that a value of a taken near +-026 will by no means bring the results on the N. and S. points , or on those approaching the N. and S. points , into accord ; and that as we have already found in the dip corrections of the Erebus , and in the declination corrections of both ships , a considerably higher value of a is required for the observations on the return from the high latitudes , than f6r those when the ship was in progress from the lower to the higher dips .
We have no observations at the Falkland Islands ( made at the spot in the ship where Mr. Fox 's apparatus was used ) either of the direction of the compass needle , or of the force acting on the horizontal needle : we must therefore obtain a and b -directly from the observations of Inclination and Intensity .
The observations gave as follows : Ship 's head .
Inclination observed .
Intensity observed .
-= -51 ? ?
56 ' .
q= 1-336 .
N. -52 46-5 -52 46-5 1-320 1-320 N.N.E. ; -52 1 }-247 }1-3151 1 N.W. -52 43 313 14 N.E. , -52 471 , 1-314 N.w. -52 45 - " 4 ' 1-312 1313 E.N.E. -52 52 -52 1-336 1-322 W.NW .
-52 38 J 1-308 E. -52 31 1-336 } w. --52 13 - ; 31-324( J,.s.r~ .
1'324 } 1'330 E.S.E. -52 16 1-355 1-350 w.s.w. --51 461 --2 01 1345 1350 S.E. ---51 321 , 1-370 s.w. -51 32 51 32 1-359 1364 S.S.E. -51 09 } 1-3681 s.s.w. --51 1 1'366 S. -50 53 -50 53 1-370 1-370 For a , we have from equation ( 1 .
) , asinOcoscos cos0coso , whence we obtain , from the obsetvations on the N. and S. points , a= +'0311 , and from those on the N.N.E. and N.N.W. , S.S.E. and S.S.W. points , a also --=+0311 .
In the Erebus we have found a for the spot in the ship where Mr. Fox 's apparatus was used ==+-023 .
from the observations made when the ship was in progress to the southward ; and = +X029 at Hobarton and the Falkland Islands .
The corresponding values in the Terror are +'026 and +-031 .
In the case of the Terror , therefore , I have employed separate tables for the corrections for the ship 's attraction , viz. a taken as +*028 in the passage from Hobarton to New Zealand ; as +*026 in the passage to the higher latitudes- ; and as +-031 during the return from the high latitudes to the Falkland Islands .
For b and c , we obtain from the observations at the Falkland Islands as followsIn the case of b , we have from equation ( 2 .
) , beos=cosO'sint'cosecg ; o2 97 the observations at N.E. , N.W. , S.E. and S.W. give b=+-984 ; those at E.N.E , ; W.N.W. , E.S.E. and W.S.W. , b=-984 ; and those at E. and W. b=*982 .
In the case of c , we have from equation ( 3 .
) , -sin ccosOcos+-dsinO ; from the observations at N. to N.E. and N.W. inclusive , and from S. to S.E , and S.W. inclusive , eliminating d , we have c= + 009 .
The constant d is perhaps the most difficult of the constants to ascertain satisfactorily , as its value derivable from the observations depends on a knowledge of the true geographical dip at the place of observation , free fi orn what is now known as station error .
Experience has fully shown the general fact , that inclinations observed on land cannot safely be assumed as free from local disturbance .
The discrepancies of gravitation at the Falkland Islands are well known from the experiments with the pendulum ; and from the geological character of these islands , we might be prepared to expect the existence of magnetic discrepancies also .
By the needles in both ships , the inclination was found a tlird of a degree higher at the magnetic observatory on shore than when observed on board in the harbour ; if the observatory dip were to be assumed as an undisturbed one , we should obtain d in both ships considerably less than unity , whereas from the comparison of the observations in both ships in the preceding December and January , with the inclination observed at the same time on the ice over a deep sea , where no local attraction can be imagined to exist , we have d ( as far as the small differences of geographical position will permit us to judge ) differing scarcely , if at all , from unity in either ship .
The preference is certainly due to the deduction from the results obtained on the ice .
Taking therefore d= 1 , c=+,1O1 b=*984 and a=+'026 , we have the corrections , and the corrected inclination , of the observations in the Terror between the 6th and 16th of January as follows : Ship 's head .
N.f l Inclinati rretion .
Correted Inclination .
observations .
observed .
N. : 4 -81 195 +i 6 ; -79 53 .
5 E* S. W. ?
s. sw ?
}6 -79 o4s5-0 051 -79 56.0 9 s. 6 -78 30-0 -1 24-5 -79 545 Slight differences in the corrected results must be looked for , as the observations were not all taken precisely at the same geographical spot : those which appear in the table are , however , very slight ; the accord produced by the corrections seems as 198 satisfactory as could be wished or expected ; and I have accordingly taken the above stated values of b , c , and d , for the whole period under notice .
On a general review of the examination to which the observations in the Erebus and Terror in this and the preceding voyage have been subjected , in reference to the magnetic influence of their iron , we find reason to conclude from the consistent experience of both voyages , that the disturbance in theim was altogether such as would be occasioned by the magnetism induced in the soft iron of the ship by the magnetism of the earth , -if we permit ourselves to include as possessing the quality of softness , certain portions of iron which , though not permanently magnetic , do still retain polarity , and require some time to conform to the changes in magnetical relations induced by changes of geographical position .
It is not improbable that this may be a general case in sailing vessels similar to the Erebus and Terror ; but we should by no means be warranted in deriving a corresponding inference in regard to ships which contain steam machinery , and still less in the case of iron vessels .
These may possibly possess permanent magnetism strictly so called ; in addition to induced magnetism , and temporarily-abiding polarity .
It is very desirable that we should have some means of judging of what may be expected in vessels of these two classes .
The knowledge would be valuable were it only for the compass corrections necessary for the ordinary purposes of navigation ; and it appears indispensable before a correct judgment can be formed of the confidence to which mnethods may be entitled , which have been already , or may hereafter be devised , to supersede these corrections by the employment of compensating forces .
It is not necessary that steam or iron-built ships should perform voyages like those of the Erebus and Terror to procure this knowledge ; a voyage from the British Channel to the Tropics would be sufficient ; the ship should be swung before her departure from these islands , and immediately on her arrival in the Trfopics , and at intervals of three or six months during her continuance there ; the experiment should also be repeated on her return to England before any material alteration is made in the distribution of her iron .
Index Correction .
Index Correction of R. F. 5 for the Observations of the Inclination in the Erebus ; The observations at sea with this needle having been made in the one position of the instrument only , viz. with the face of the circle towards the east , and the marked side of the needle towards the observer , we have to obtain the index correction , by comparing the inclinations observed in the same manner on shore , ol on the ice , with the results given at the same places by needles of which the poles were reversed and the needle and circle used in the eight ordinary positions .
The stations which furnish this comparison are Hobarton , Sydney , New Zealand , the Falkland Islands , and two stations on the ice in the latitudes of -63 ? ?
23 ' and -65 ? ?
49 ' .
The results of the observations at Hobarton with needles with which the complete process for determining the inclination was gone through , were given in No. V. of these Contributions* .
Those at the other five stations are as follows:* Philosophical Transactions , 1843 , Part II .
p. 165 .
99 From these observations we have the true inclination at these six stations as follows:0 t On ice , lat.-65 49 .
Long. On ice , lat.-63 23 .
Long. Hobarton ... ... Sydney ... ... .
New Zealand ... . .
Falkland Islands ... .
0 202 210 I0 02 .
79 02 .
77 ... .
70 ... .
62.* ... 59 ... .
52 The observations with R. F. 5 , at the same stations , and at the same spots on shore , or on the ice , gave as follows : On Ice .
Lat. -650 49 ' .
Long. 202 ?
02 ' .
0X Face East -79 35'6 Face West -80 39-2 Mean -80 07'4 On Ice .
Lat. -630 23 ' .
Long. 210 ?
02 ' .
Hobarton .
-77 15-5 -70 26.4 -78 20-3 -71 20-3 -77 47-9 -70 53-4 Sydney .
0 -62 46'3 -63 44'3 -63 15-3 ?
~ ~__L New Zealand .
Falkland Islands .
-59 29'8 -52 32'9 -60 279 --53 34'7 -59 58'8 -53 03'8 We have thus the following index corrections : Face East 3'8 Face West +59-7 Mean correction +27'9 ii 7-8 +57-0 +24'6 14'3 +39'6 +12-7 2'8 +55-2 +26-2 21 +56-0 +26'9 + 6-7 +68-5 +37.6 t , and the difference of the results with the face east and face west as follows:63'6 64'8 53.9 58'0 58I1 58'1 61 *8 From the signs and numerical values of the corrections of the mean results with R. F. 5 , we may infer that the axis of rotation in this needle ( leviated firo the centre of gravity in the longitudinal direction , so as to cause the south end of the needle slightly to preponderate .
From the differences of the results with the face east and face west , it appears that there wwas also a small deviation in the axis of rotation from the centre of gravity in the perpendicular direction .
In the results with the face east , these two sources of error partially counteracted each other , so that the index correction with the face east amounted at no time to more than a very few minutes .
The corrections which have been applied to the observations have been taken fionk the following table , in which the correction for -70 ?
has been taken as --5'8 , and the change in the correction , corresponding to an increase of one degree in the south dip , as -0'5 .
In forming this table the determinations on land have been allowed a greater weight than the determinations upon the ice , the latter consisting of fewer observations , and being made probably under circumstances less favourable for this particular purpose .
39.5 23'3 40-7 49'1 31'9 26-2 Table of Index corrections for R. F. 5 , face East , between --52 ?
and -85 ? .
Inclination .
Correction .
Inclination .
Correction .
| -52 +3.2 -69 5.3 -53 +237 -70 5'8 -54 +2 22 -71 6-3 --55 +1*7 --72 6'8 -56 +1-2 -73 73 -57 +0'7 -74 78 -58 +0*2 -75 8-3 -59 --03 -76 88 -60 -0.8 -77 9-3 -61 -1-3 -78 9.8 -62 -1-8 -79 -10-3 -63 --2-3 -80 -10-8 -64 -2-8 -81 -1 13 -65 -33 -82 -11.8 -66 -38 -83 -12-3 -67 -4.3 -84 -12'8 -68 -4-8 -85 -133 i 105 Index ' Correction of F. C. B.for the Observations of Inclination in the Terror.-The observations of inclination at sea in this ship were all made with the face of the instrument towards the east , and with the marked face of the needle towards the observer .
We may examine the index corrections consequently in the same manner , and by comparison with the same complete determinations as in the case of the needle of the Erebus ; confining the comparison however to the land stations , because F. C. B. was not observed with at either of the ice stations .
The inclinations taken with this needle were observed both direct and with the aid of deflectors ; the deflectors employed were a spare needle as " deflector N " and " ' deflector S " ; and the magnets of the apparatus , either used separately as " magnet N , " or " magnet S , " or conjointly as " magnets N S. " From some instrumental accident , the inclinations observed with " deflector N " were always considerably in defect of the others when the face of the circle was east ; with a corresponding excess with the face west , on the few occasions on shore when the observations were made in both positions .
As the observations at sea were exclusively with the face east , it has been necessary on this account to consider separately those amongst them which were taken with " deflector N , " and to obtain a distinct index correction for them .
We will first examine the index corrections required for the direct observations , and for those with the other deflectors .
The observations with F. C. B. on shore at the four land stations , where the conm , plete process for determining the true inclination was needles , were as follows : Hobarton .
Face East -70 17-3 Obset ved O. [ Obseirved Face West -70 44'8 Mean ... .
-70 31-1 True inclination -70 40'7 FFace East -23'4 Index correction Face West + 4'1 [ Mean. .
9-7 Differences face East and West 27'5 The corrections of the mean results with gone through with other Sydney .
New Zealand .
Falkland Islands .
O , 0/ 01 -62 22*4 -58 50-6 -51 38'4 -62 565 --60 02'8 -52 57'2 -62 39'5 -59 26-7 -52 17'8 -62 491 -59 31'9 -52 26'3 -26'7 -41'3 -47-9 + 7'4 +309 +30-9 96 52 -8-5 34'1 72*2 78'8 F. C. B. at the four stations accord well within the limits of observation error .
On examining the differences in the results with the face east and face west , and the corrections severally required in the two positions at the four stations , it appears probable that a very slight derangement of some part of the instrument took place between the observations at Sydney and those at the Bay of Islands , which caused the partial results with the face east and face west to diverge more from each other than they had done previously , but without affecting the mean results .
A note which accompanied the observations to England shows that Captain CROZIER considered that some slight change had taken place in the amount of the index correction with the face east , but was unable to assign its date or its cause .
In the absence of any distinct evidence in these respects , and in consideration of the insufficiency of the means of assigning the precise amount of the change , -I have preferred the employment of an arithmetical mean of the index corrections observed at the four stations ( -35 ' ) during the whole course of the voyage .
The uncertainty arising from this source cannot amount to more than a very few minutes in any portion of the voyage .
For the index correction with deflector N we have , Hobarton .
Sydney .
New Zealand .
Falkland Islands .
Face East ... -69 33'5 -61 36-7 --57 58'0 -50 54.4 Face West. .
-71 25'9 -63 007 -60 12'3 -53 31'3 Mean ... . .
-70 29'7 -62 18'7 -59 05-1 -52 12'8 True inclination-70 40-7 -62 4941 -59 31'9 -52 26'3 Index correction , face East. .
-67-2 -724 --93-9 -91*9 1Mean .
index correction face East ... ... .
Mean index correction , face East ... ... ... .
--81 ' 106 Elements of Calculation of the Intensity Observations .
1 .
With Weights.-The observations of the intensity of the magnetic force , during the period now under consideration , were made in both ships with Mr. Fox 's apparatus ; those in the Erebus with the same circle which had been used in the previous voyage , and those in the Terror with a circle of the same size as that of the Erebus , being the property of Captain CROZIER , and received by him at Van Diemen Island .
The needle employed to show the angles of deflection in the Erebus , marked R. F. 5 , was not the same which had been used for that purpose in the voyage of 1840-1841 , namely , R. F. 4 , which now in its turn was used as a deflector .
The weights employed in deflecting the intensity needle were 1 , 2 , 3 , 4 , 5 and 6 grains : the angles of deflection obtained with otne grain were however too small to yield results of the same satisfactory nature as those derived from the weights from two to six grains , and I have not therefore taken them into the account .
The mounted needle in the Terror was marked F. C. B. , a spare needle C being used as a deflector , in addition to the deflecting magnets belonging to the apparatus .
The weights were 1 , 1 , 2 , 2- , 3 and 32 grains .
At Hobarton we have the deflections occasioned by the constant weights on the needle of the Erebus , April 1841 , as follows : Deflection .
Therm. Deflection .
Therm. grs. 0/ 2 13 02.8 60 3 19 37'2 60 Face East.{ 4 26 47'7 60 5 34 23'5 60 642 55-7 61 and in the needle of the Terror as follows : Deflection .
Therm. grs.,1 I12 Face East .
&lt; .
21 23 31 0l 12 11'9 18 29'4 25 13'7 31 43'0 39 02'3 46 51'3 0 60 60 60 60 60 60 grs. r2 13 14*5 60 3 19 55'5 60 West .
&lt; 4 27 027 58 5 34 51'5 58 L6 43 07'3 58 Face West .
&lt; Deflection .
Therm. grs. 01 11 42.0 60 117 52-6 60 2 24 15*6 60 ' 22 31 00-7 60 3 38 42'3 60,346 06'3 60 At Sydney , in July 1841 , the deflections with the same weights wereEREBUS .
Deflection .
There .
13 57-4 56 2 21 13-7 55 3 29 09'2 55 &lt; 4 37 43-3 55 i5 46 51-7 55 -6 Deflection .
There .
0I0 14 32'6 64 21 51'4 63 29 32*1 64 37 38-9 63 47 32-4 63 TERROR .
Deflection .
There .
grs. Oo '1 13 08'8 60 11 20 02'0 60 2 27 007 60 | 234 25-2 60 |3 42 06'9 60.3 51 1355 60 grs. { 2 .
3 cd , 4:4 56 grs. 1112 2.2 32 Deflection .
0l 12 44-1 19 033 26 01*2 : 33 17'7 41 35-2 51 021 There .
0 60 60 60 60 60 60 107 t Taking 1*82 as the provisional value of the intensity at Hobarton ( Phil. Trans. 1843 , Part II .
p. 186)* , we have its value at Sydney , by the needles of the two ships , as follows : EREBUS .
Face East .
Face West .
1-703 1P662 1'687 1'667 1*683 1*680 1'680 11704 1'698 1-688 1-690 1'680 1 685 = grs. 1 1l 2223 31 2f TERROR .
Face East .
1'691 1-685 1'708 1 692 1.709 1 703 1'698 1*699 At the Bay of Islands in New Zealand , in August and October 1841 , the deflections were as follows : EREBUS .
Deflection .
There .
-15 23'3 60 23 17'9 59 30 26'9 59 40 52'0 60.51 26'0 61 Deflection .
rl4 43.2 1 22 45*0 CI &lt; 30 30-6 | 39 59'3 50 35.0 October .
There .
Deflection .
00 t 68 J15 11l1 70 1 23 17-2 70 31 29-2 70 d 40 51*0 71 51 38*7 TERROR .
Deflection .
There .
14 03'2 21 17'9 28 22-1 36 50'7 44 58'3 55 09'9 5 59 59 59 59 59 59 August 4 .
) ai 'I C ) &amp; O a ) i CS Deflection .
There .
F13 24'359 20 30 5 59 27 46-9 59 35 43'0 59 44 38-7 59 s55 23*7 59 4 .
; ) Cd r Cd CS Deflection .
0I 13 51-7 20 53 0 28 22'4 37 05-6 45 02*2 s55 19'1 October .
There .
Deflection .
64 " 13 26 8 64 : 20 16-4 64 27 38-8 64 g 35 45'1 64 44 47-7 64,.55 26-4 whence we have the intensity at the Bay of Islands , by the as follows:needles of the two ships , * 1'82 +e bging the true value , in which e is a small correction to be determined hereafter , applicable to the whole series of 9bseryations depending on Hobarton as a primary station .
grs. 2356 Face West .
1-674 1-712 1-705 1-709 1.715 1-687 1-700 grs. f2 &amp; 3 6C 145 1 C3 5.6 August .
r C ) C. ) PI Deflection .
ot 14 59.3 22 47'5 30 55'0 40 10'5 50 38*1 There .
59 59 59 58 58 There .
64 64 65 -65 65 4 .
; C , 0 ) c grs..1 I1 2I2 22 3 23 There .
0 64 64 64 64 64 64 108 Besides the four land stations at which the intensities shown by the needles of the two ships have been thus compared , we have also one ice station in lat. -65 ? ?
47 ' , long .
202 ?
08 ' , at which similar comparisons may be instituted .
The deflections and intensities were as follows : EREBUS .
TERROR .
Deflection .
There .
Intensity .
Deflection .
There .
Intensity .
grs. oI grs. oo r2 12 130 50 1940 11 1254 53 1P940 11 3 18 32'4 54 1-921 1 17 08'3 53 1-957 SC 4 24 49,3 54 1952 2 23 02'9 53 1-979 5 32 02'4 54 1P936 fl 229 16'2 53 1P955 C6 39 31-4 55 1P946 3 36 17-4 53 1P935.939 L31 43 23'5 53 1'932 1'950 Collecting these several results in one view , we have as follows : EREBUS .
TERROR .
DIFFERENCE .
Intensity at Hobarton ... ... 182 1'82 ( Erebus in defect .
) Intensity at Sydney ... ... .
1685 1699 '014 or 8 parts in 1000 Intensity at the Bay of Islands ... .
1-594 1-607 '013 or 8 parts in 1000 Intensity on ice , lat. -65 ? ?
49 ' , long .
202 ?
02 ' 1'939 1 950 01 1 or 7 parts in 1000 Intensity at Port Louis , Falkland Islands .
1'322 1'336 '014 or 10 parts in 1000 The difference between the results given by the needles of the two ships , though small , is so consistently shown at all the stations during the voyage , that we cannot hesitate to attribute it to the occurrence of a change of corresponding amount in the magnetism of one needle or the other , between the observations at Hobarton in April 1841 , and those at Sydney in July of the same year .
If we further compare the intensities observed at sea by the two ships on the passage from Hobarton to Sydney , we find that a similar difference prevails in them ; and we are therefore led to the conclusion , either that the needle of the Terror gained , or that the needle of the Erebus lost , a very small portion of magnetism , in the period between the observations at Hobarton in April 1841 , and the departure of the Expedition from that port in the following July .
Now experience has shown that a loss of magnetism is no unfrequent occurrence , whilst a gain is extremely rare , happening only , as far as we know , from such an accident as the contact of a needle with a more powerful magnet than itself .
We may therefore conclude with great probability that the needle of the Erebus sustained a small loss of magnetism between April and July 1841 , antecedent to all the observations of the voyage , causing the intensities derived with it , when computed in reference to the angles of deflection observed at Hobarton in April 1841 , to require to be increased about one hundredth part , or more precisely 8 parts in 1000 , in order 110 to bring them into strict relation with 1 82 , taken as the value of the force at Hobarton .
This correction being applied , all the intensities observed throughout the voyage by the two ships are in accordance ( subject only to errors of observation ) , forming a consistent series of relative determinations , resting on 1'82 and 1*336 , assumed provisionally as the values of the intensity at Hobarton and Port Louis , the cormmencing and concluding stations of the series .
The correction is made in the Table which exhibits the intensities observed on board the two ships , and the geographical positions to which they belong ; it is also made in the results inserted in the Map .
The correctness of the values assumed at the base stations , 1*82 at Hobarton and 1P336 at Port Louis , remains to be proved by absolute determinations which have yet to be made at those two stations .
The absolute intensities observed by the Expedition itself , with the instruments and according to the method prescribed nh the instructions of the Royal Society , certainly have not the necessary precision .
In the preceding Number of these Contributions are stated the results of five determinations which were obtained by Captain Ross at Hobarton in 1840 and 1841 , with the 15-inch magnets of his observatory magnetometers ; and of twenty-two determinations obtained by Lieut. KAY at the magnetic observatory at that station , with similar instruments , in 1841 and 1842 .
Captain Ross 's mean result was 4'573 , the partial results varying from 4'491 to 4'626 .
Lieut. KAY'S mean result in 1841 was 4'553 , the partial results ( ten in number ) varying from 4'509 to 4'601 ; and in 1842 4'513 , the partial results ( twelve in number ) varying from 4*443 to 4*568 .
In 1843 Lieut. KAY received the auxiliary apparatus supplied in compliance with the revised instructions of the Royal Society , published in 1842 .
The magnets of this apparatus were 12 inches in length .
The following Table exhibits the results obtained with this instrument in thirteen determinations made with it , between June 23rd and July 1st , 1843 .
Each determination is deduced from two series of observations of deflection ; in the first six instances the distances were 4*505 and 6*005 feet ; in the remainder , 4-0 and 5*3 feet .
The moment of inertia of the deflecting magnet was computed from the length , breadth and mass of the bar .
June 23 .
4'509 June 27 .
4*557 24 .
4*515 28 .
4*505 24 .
4'528 28 .
4'504 26 .
4*510 29 .
4'549 26 .
4-523 29 .
4*527 27 .
4'583 30 .
4'466 July 1 .
4'479 Mean of the 13 determinations 4*520 Here also it is obvious , from the discrepancy of the partial results , that the angles of deflection afforded by these magnets at the prescribed distances , viz. the least distance being not less than four times the length of the bar , were still too small ; and that before any final conclusion be arrived at , it is desirable that we should await the results which will be obtained with the smaller apparatuls described by Lieut. RIDDELL in his " Magnetical Instructions for the use of Portable Instruments , " &amp;c. In this apparatus the suspended and deflecting magnets are respectively 3-0 and 3-67 inches in length .
Meanwhile we may derive , as a provisional value , the arithmetical mean of the four mean results already stated ; allowing to each an equal weight , we have , Ross , in 1840-1841 , 15-inch magnets , 4*573 KAY , in 1841 , do .
do .
4'553 KAY , in 1842 , do .
do .
4513 ean KAY , in 1843 , 12-inch magnets , 4-520 which , with the other necessary data stated in the preceding Number of these Contributions , would give the value of the total intensity at Hobarton 1-81 to 1P372 in London* .
Since these pages were written I have received the details of the observations of ten distinct determinations of the absolute horizontal intensity at the magnetic observatory at Hobarton , made in August 1843 with deflecting and suspended magnets respectively of 9'18 inches and 7*50 inches in length .
The deflecting distances were the same throughout , being 3'2893 and 4'3393 feet .
The calculation of these observations not having been yet received from Lieut. KAY , the results have been computed by Lieut. RIDDELL , R.A. , F.R.S. , so far as the materials hitherto furnished permit .
They give the value of X ' ; -being the absolute horizontal intensity ( X ) , uncorrected for the difference in the magnetic moment of the deflecting bar produced by the earth 's inducing action in the different positions in which the bar is placed in the experiments of deflection and in those of vibration ; viz. 1 ?
perpendicular to the magnetic meridian , and 2 ?
in the plane of the meridian .
We owe the suggestion of a correction due to this cause to Dr. LAMONT : but the necessary data for computing it , for the particular bars employed by Lieut. KAY on this , or on the former occasions , have not yet been received .
Observations made at the Cape of Good Hope and at Woolwich , with similar bars , have given results which show that the correction may possibly prove to be of nearly the same amount for the larger and smaller bars , in which case the relative values will be but little affected , and we may estimate that the value of X at Hobarton will be about 0'02 less than X ' .
In the expression which has been employed in these Contributions for the absolute horizontal intensity ( 1P82-+e at Hobarton and 372 +e at London , e being a small quantity to be supplied hereafter ) , the correction here referred to will form a portion of e. The following Table exhibits the abstract of the observations made in August 1843 with 9'18 and 7'50 inch bars .
Deflecting Magnet .
Bifilar Magnetometer .
Values k=000229 .
q-000224 .
Gottingen Mean Time .
Temperature aes 000229 000224 .
No. Value during o. .
of m ' .
deflection .
Reading .
Temp. dh 1843 .
Aug. 20 19-0 9-18 inch .
6-256 54-6 4-5052 165*1 52-0 21 11-5 9-18 inch .
-259 49-6 -5034 ]68-6 49-1 21 16-5 9-18 inch .
-251 51-9 -5043 165-3 49*1 21 19-5 9-18inch -261 53-7 -4993 168-3 50-0 22 110 9-18 inch .
-227 48-0 *5177 165-4 49-3 22 19-5 9-18inch .
-243 54-5 -5025 164'6 50*7 23 10'8 9*18 inch .
*259 50-7 -4884 161-0 51*2 23 18-1 9-18 inch .
-24452*4 5005 162 2 51-0 23 19-1 9-18 inch .
-240 5230 -4982 163-9 51-3 25 11-4 9-18 inch .
-252 49-4 *4953 165*3 51'5 6-249 51-7 4-5015 165-0 50-5 The mean value of the results , 4-501 , is considerably different from the mean deduced in the text from all At the Falkland Islands there were two determinations of the absolute horizontal intensity made by Captain Ross at the Magnetic Observatory at Port Louis , one in September 1842 , being 6'87 , and a second in November of the same year , being 6'32 .
They were both made with 15-inch magnets ; the angles of deflection were observed at four distances , but amounted only to 56'"8 , 31'-9 , 21'-4 , and 12 ' 9 in the first experiment , and to 1 ? ?
49''9 , 1 ?
01 ' 6 , 41 ' 5 , and 25 ' 1 in the second experiment .
These values of the horizontal intensity would give that of the total intensity at Port Louis respectively 1*609 and 1P367 .
It is obvious that we can draw no conclusion whatsoever from these numbers , and that we must wait for the confirmation or correction of the value given by the needles of Mr. Fox 's instrument , until absolute determinations can be proctured with instruments capable of affording more satisfactory results .
Steps have been taken to obtain such determinations at the Falkland Islands from Captain SULLIVAN , R.N. , and at Sydney and New Zealand from the Surveying Expedition under Captain BLACKWOOD , R.N. ; when these arrive , we may learn whether any and what final correction will irequire to be applied to the intensities now provisionally deduced from the observations with Mr. Fox 's needles , in the Erebus and Terror .
We may expect to receive these determinations before the time when the results now presented to the Royal Society will have to be combined with those of the preceding and succeeding years , in a general calculation of the magnetic lines in the southern hemisphere .
2 .
With Deflector &amp; s. In the Erebus , the spare needle R. F. 4 was employed , -as c deflector S , " with its south pole opposite to the division of the circle which the south pole of the mounted needle had previously indicated as the dip ; -and as " deflector N , " with its north pole similarly applied to the opposite division of the circle .
The angles of deflection varied in different localities during the voyage , in round numbers as follows:-Deflect .
S from 52 ?
to 71 ?
; and deflect .
N from 49 ?
to 67 ? .
For obtaining the equivalent weights to the deflecting force of the deflectors at these angles , we have the comparative observations with deflectors and weights at Hobarton , Sydney , New Zealand , the Falkland Islands , and on the ice in lat. -65 ? ?
47 ' , long .
202 ?
08 ' .
The angles of deflection caused by the weights have been already stated ; the preceding observations ; yet from the improvement which it is natural to suppose practice must have made in the observers , and from the reduced discrepancies of the partial results with the smaller bars , the mean of the ten results in August 1843 would seem entitled to a preference over the earlier and more numerous results .
Judging by what has been done at Woolwich with the 2*45 and 3 inch magnets , and at the Cape of Good Hope with 3-0 and 3 67 inch , we may expect with them a still further and considerable reduction in the discrepancies of the partial results ; but it would not be safe , with the comparisons which we have now before us , to feel full confidence that there will be no apparently constant or systematic difference between the results of the larger and smaller bars .
Reviewing the whole subject , we can as yet , therefore , only consider ourselves as being in progress towards such accuracy in determining the ratio of the intensity at different places by the ab. .
solute method , as shall be superior to that with which it was previously obtained by the employment of wellselected needles in relative determinations .
Q2 113 those by the deflectors , with the equivalent weights deduced from the comparison , are collected in the following Table .
t Angles of deflection Equivalent Intensity by weights .
Station .
Date .
deduced by weights .
weights .
Def. S. Def. N. Def. S. Def. N. vr o.__grs .
grs. Hobarton ... . .
April 1841 ... ... 182 286 86 53 0'26 7 39 7 08 Sydney ... ... .
July 1841 ... ... . .
1685 59 10-2 55 37-0 7-05 6-77 New Zealand ... .
Aug. and Oct. 1841 1'594 61 46-9 57 59-0 6-84 658 On ice ... January 1842 ... .
1-939 54 03-1 50 35'0 7'65 7-30 Falkland Islands Apriland Aug. 1842 1322 71 11'8 67 10'3 6-10 5993 By projecting these angles and weights , and proceeding in the manner described in the Third Number of these Contributions " , the values of w ' in the following Table were obtained for each deflector , corresponding to each angle of deflection v ' ; and employing these values of w ' , the intensities 1 ' entered in the general table of observations have been compited by the formula 1 82s in 56 28-6 182 sn 56 wcosec v'-2'053w ' cosec vt 7'39 Besides the observations with the spare needle R. F. 4 , employed as a deflector , angles of deflection were occasionally observed with the magnets N and S , belonging to the apparatus of the Erebus , used conjointly ; their magnetism , however , was so much inferior to that of R. F. 4 , that , even when both were used together , their joint effect was less than the half of either pole of II .
F. 4 ; their results would consequently be much inferior in precision to those of R. F. 4 , and I have not therefore employed them .
Def. S. Def. N. H ' .
- ' : 1 ' .
tat .
W ' ' W/ ' .
' .
: W 1 .
v. W ' .
w V. grs. gr. grs. grs. g : rs .
52 7-87 59 7-11 66 647 49 749 656 6'76 63 619 53 7-76 60 701 67 639 50 7'38 57 667 64 6-13 54 7.65 61 6*91 68 6'31 51 727 58 6-57 65 6-06 55 7-54 62 68s2 69 6-24 52 7-17 59 6'48 66 6-00 56 7*43 63 6-73 70 6-17 53 7-07 60 6.40 67 5-94 57 7.3 64 664 71 i 6-10 54 697 61 6-33 58 7'21 65 6-55 72 6'03 55 686 62 6 26 In the Terror , the spare needle marked C was employed both as " deflector N " and " deflector S. " The magnets belonging to the apparatus were also used , N separately , and N and S conjointly .
Observations were also occasionally made with magnet S , but its magnetism was so feeble , and the deflections obtained with it consequently so small in comparison with the others , that the results are not entitled to the same confidence , and have not therefore been taken into the account .
The equivalent weights have been obtained , as in the Erebus , from the comparative observations with weights and deflectors at Hobarton , Sydney , New Zealand , the Falkland * Philosophical Transactions , 1842 , Art .
II .
114 Islands , and on the ice in lat. -65 ? ?
47 ' , long .
202 ?
08 ' .
I have also , in the case of the Terror , availed myself of a comparison of the weights and deflectors made on the 3rd , 4th and 5th of December 1841 , at sea , when the weather was extremely favotwrable , and the ship did not materially change her position .
From the observations on these days we have as follows : December intensity Angles of deflection by e1841 .
deduced..,. .
by weights .
Def. N. Def. S. .
N. Ma .
$ .
, .
r5 r , i 1. .
XX 3A.M . !
1'783 ; 6 55'9i 34 0'6'7 30 44'1 40 52-8 3 P , M. 1-778 36 51'6 34 06330 46-1 40 45'8 4 1'773 36 4478 34 22-0 30 48-7 40 56'3 5 1-779 36 18'3 34 29-4 30 46'1 40 54'9 Mean. .
1-778 36 42.5 34 16-0 30 46-2 40 52'5 The several comparisons from which the equivalent weights are derived , together with the weights so derived , are collected in the following Table .
Inten ?
Angles of deflections by Equivalent weights .
Da e1 sity de ... ... ... ... ... ... ... .
Station .
Date , .
: duced by eDeflector Defector Magnet Magnets Deflector Deflector Mt Magnetgnets i weighls , N. S. N. N S. N. S. NS .
10 ; om o , o/ 1 grs. grs. grs. grs. Hobarton ... April 1841. .
: 1'820 36 0'06 33 23 .
0 30 14-0 40 05'5 2*793 2.613 2-391 3-059 Sydney ... .
July 1841. .
1'699 38 05-9 35 15-7 31 4721 41 45-3 2736 | 2560 2336 2*953 New Zealand. .
Adc 1.608 63 368 36 57-8 32 50-8 42 58'4 2'675 2^525 2276 2-861 Dec.3,4and 73 '5 3 At Sea ... ... 11841 ... i 1 778 36 42.5 34 16-0 30 46'40 52-5 ' 2-773 2-613 12374 3-036 One. .
Jan. 16 , 1842 194933 476 31 16-1 28 52-7 38 45-7 ' 2829 2 640 2'456 3-184 Fn u Isands prlJ uly andi 14 31 FalkladIsland1April , J 1336 44 38-2 41 57'1 35 59'0 46 14-0 ?
'442 2'324 2-042 i2'510 Aug. 18'42 5 The equivalent weights for each deflector , and for each half degree of deflection , have been obtained in the manner already described , for the angles of deflection and equivalent weights in the preceding Table , and are subjoined ; by their aid the intensities I ' entered in the general table of observations have been computed by the formula I-='383u2w cosec v ' .
Def. N. Def. S. 1 Magnet N. .
Magnets NS .
v w. UI ?
, .
I v.I wI IW.1 I I , w ' .
w. 1 v ' .
IJ W " .
It^. .
t. v. WwI Ig is grs. fgrs .
gV .
0 1 .
-I gs _ , ,g .
' gs Vg rs .
v 't_ , g _ -33 00 2-840 39 30 2-682131 00 2-644 37 30 2-504 8 001 2-482 34 30 2-166 37 00 3-268 43 30 2-816.33 30 2-834 140 001 2660'31 30 2-639 38 00 2-485 28 30 2-464135 00 2-126 37 30 3-240 44 00 2-766 '34 00 2826,40 30 2638"32 00 2-63438 30 2-464 29 00 2446 35 30,2-085 38 00 3-210 44 30 2-714 34 30 2-817 '41 00 2615 -i32 30 39 00 2-444 29 30.426 00 ' 2040 38 30 3-180 45 00 2-660 35 00 ; 8091,41 30 ' 2-593133 00 2-619 39 30 2'423 30 00 2-406 39 00 3'14845 301 2604 35 30 ' 2-800'42 00 ; 2571'i3 30 2611 040 00 2-403 30 30,39 30 ' 311446 00 2544 36 00 2-790 42 302548 34 00 2-602 40 30 2'382 31 00j 2'367 ' 40 00 3-081 46 30Q 2'460 36 30 2-777,43 00 2'524 34 30 2-591 41 00 2361131 30i 2-346 ; 140 301 30499 37 00 2-765 43 30 ' 2488 35 00 2-580 41 30 , 2-34132 00 ? ?
232 3 41 00 3-016 8 ?
341i ; 32 QO ~2 410 37 30 2-753 44 00 2-473 35 30 2-567 42 00 2-321'32 30 2-297 41 30 2-981 38 00 i 2-738-44 30 2-448 36 00 2-554 33 00 2 i268 42 00 2-944 38 30 2-721 -45 00 2-423 36 301 2538 33 30 2236 42 30 2-905 ' 39 002'2702 137 001 2-523 '34 00 2'203 43 00 2'863 , ~_. .
ll II ; C--III _ ?
II1l ~~I.---lli I-l-r II ?
L~3L~LI -II* LIE l*IIl !
36---'- ?
--- ; 11 5 General Remarks.-If we take a general view of the magnetic DECLINATION in the southern hemisphere , particularly in the best-known portion of it , comprised between the tropics and the Antarctic Circle , we find that the phenomena present the same obvious and decided features of a duplicate system as do those of the northlern hemisphere .
If , following any of the geographical parallels , we carry our attention round the hemisphere , we find it divided into four spaces , in which opposite characteristics in regard to the direction of the needle alternately present themselves .
In two of the spaces the change in the pointing of the needle , as the space is traversed in the direction of the parallel , is continuous and progressive towards the west , and in the other two continuous and progressive towards the east .
If , for example , commencing with the meridian of 30 ?
E. or thereabouts , we trace the parallel of -45 ?
round the hemisphere , always proceeding in an easterly direction till we return to the meridian at which we began , we shall find that we first pass through a space in which the direction of the north end of the needle becomes progressively more and more easterly , eithler by the decrease of westerly or increase of easterly declination ; we next pass into a second space , on entering which the continuity is broken , the progressive movement of the north end of the needle towards the east is arrested , and its direction becomes now more and more westerly as we advance ; thence we pass , successively , into a third space which has the same characteristic as the first , and into a fourth which has the same as the second .
The spaces here spoken of must be distinguished from those which are characterized by the exclusive prevalence of either east or west declination : they have a more simple and pure magnetical relation , implying the predominance within each space of one or the other of the two systems of magnetic forces which govern the direction of the needle .
It may happen , or it may not happen , that in one of these spaces the direction of the needle may coincide in some point or points with the geographical meridian ; when this occurs , the space will comprise both east and west declination ; when it does not happen , the declination throughout the space will be exclusively east or exclusively west as the instance may be but in either case , the change in the direction of the needle is always continuous and uniform in character throughout the space .
It is well known that if the magnetic declination be computed on the supposition of a single central magnetic axis , there will be found two , and only two such spaces in each hemisphere .
The systematic discordance which the declinations in the northern hemisphere presented when compared with the declinations so computed , and their agreement with the phenomena deducible from a double system of forces , led HALLEY to embrace the latter hypothesis .
The declinations in the southern hemisphere present an arrangement strictly analogous to that in the northern , and conduct to the same conclusion , be that conclusion what it may .
If , with HALLEY , we view the declinations in the Southern Pacific as principally influenced by the weaker system of forces , or by that to which is also to be ascribed the high intensity of the magnetic force in the same quarter , we should be prepared 116 to expect that if the geographical limits of the adjacent spaces , having the chlaracteristics referred to , were determined at different epochs , the alteration in the position of the spaces , if any , would show the existence ?
of a secular change in the system itself ; that it would indicate the direction of such change ; and , if the intervals were sufficiently long in reference to the precision with which the determinations were made , the average rate of the movement of translation might also be inferred .
In this view a knowledge of the geographical position of the limiting lines , or of lines drawn so as to separate one of these spaces from the next , may have a particular value .
In the part of the Pacific Ocean which is now referred to , the separating lines , as for distinction they may be called , coincide nearly in direction with geographical meridians , and are therefore crossed nearly at right angles by vessels pursuing a course from east to west , or from west to east .
Prior to our own times , the epoch of Captain COOK'S voyages is perhaps that in which the observations of the declination in the Southern Pacific may be regarded with the most confidence .
The determinations of that period have been collected by M. HANSTEEN into a map , of which he assigns the year 1770 as the mean epoch .
It is one of those published in the Atlas of the Magnetismus der Erde , and comprehends the results obtained by BYRON , CARTERET , WALLIS , COOK in three voyages , EKEBERG also in thr1ee voyages , and ABERCROMBIE .
If in this map we draw lines separating the spaces which have the opposite magnetic characteristics referred to , and compare them with the corresponding lines which we may draw in ERMAN'S map of the Declination in 1827-1830 , published in the Magnetic Instructions of the Royal Society , we find an effect of secular change very distinctly shown in the altered position of the separating lines .
These lines , A and B , are drawn in the accompanying Plate * , where the two epochs , 1770 , and 1827-1830 , aie brought into comparison .
In the map of 1827-1830 , the separating lines occupy a considerably more westerly position than in the earlier map , the difference amounting to about 10 ?
of longitude .
Hence we are led to the conclusion , that the spaces in the Southern Pacific , distinguished by certain magnetic characteristics , undergo a movement of translation , of which the general direction is from east to west .
This direction is the opposite to that in which the change is known to take place in the corresponding quarter in the northern hemisphere ( viz. in the Siberian quarter ) , where the secular movement is from west to east .
We are not without earlier , though possibly it may be supposed less precise , evidence of the effect of secular change in the Southern Pacific .
From HALLEY'S chart of the variation lines for 1700 , we are enabled to draw the separating line B for that epoch , when we find it to have been between the longitudes of 305 ?
and 310 ? .
In a still earlier map drawn by HANSTEEN for the year 1600 ( Magnetismus der Erde , Atlas , No. 1 ) , representing the observations of the very able and scientific navigators of that period , we find the position of the same line to have been about 333 ?
of east longitude .
In the observations of Captain Ross 's voyage , we have the most recent evidence of the progressive westerly movement of the magnetic phenomena in the Southern Pacific .
The separating lines A and B , deducible from the observations in 1842 , are seen in the Plate to be in both cases considerably to the west of those derived from the observations of 1827-1830 .
The whole body of evidence thielefore , from the earliest observations to the latest , is consistent in showing a progressive movenlent to the westward of the spaces in the Southern Pacific , characterized by certain magnetic peculiarities , which in HALLEY'S view indicated the proximity and predominance of the weaker system of forces .
It is worthy of notice that the rate of progression , deduced from the changes of position shown at the several epochs , differs much less fiomt a uniform rate than might have been anticipated from the nature of the evidence we possess , even supposing the actual rate to have been uniform in nature ; whilst the magnitude of the whole change which appears to have taken place since the phenomenon became the subject of observation , in round numbers 50 ?
of longitude in two centuries and a half , can scarcely fail to fix the attention .
These are facts which , when the true physical causes of the magnetism of the globe shall occupy the earnest attention of philosophers , will probably attain an importance which at present perhaps we scarcely sufficiently estimate .
But an endeavour to place distinctly before our minds facts of which the explanation must be deemed an essential condition of a satisfactory solution of this great problem , may not be without its use even at the present time .
It may be also useful to call the attention of navigators to the value which may hereafter attach to determinations which may be made with instruments which are on board every ship , and in constant employ for the ordinary purposes of navigation .
The position of the lines separating the spaces which have been the subject of discussion , has the advantage of being even more easily determined by observations on board ship than that decreasing , will then begin to increase , and if previously increasing will begin to decrease ; the determination is therefore independent of compass error , which is a much more prevalent source of error than is generally supposed ; and if the ship 's course be steady for some days together , which in the latitudes in question is very frequently the case , it is also in a great measure independent of the disturbance occasioned by the ship 's iron .
A very cursory inspection of the general table of the declinations observed by the Erebus and Terror suffices to show that they must have crossed the separating line ( A ) about the 15th of March 1842 , when their latitude was about -59 ?
and longitude 221 ?
; and the line ( B ) about the 27th or 28th of the same month in latitude about -59 ?
, and longitude 275 .
Shtould the circumstance occur that one of the separating lines in the course of its progressive change of place should pass over a magnetic observatory , the epoch of its passage would be precisely determined .
There is some reason for believing that * The line A passes through the culminating points of the southerly inflexion of the declination lines , of which the present position is shown in the Declination Map at the close of this paper to be about 220 ?
east longitude .
The line B passes through the culminating points of the northerly inflexion of the declination lines about the longitude of 276 ? .
118 such an event is now taking place at the Cape of Good Hope .
If we examine ERMAN'S map of the Declination in 1827-1830 , published in the magnetic instructions of the Royal Society , we find one of the separating lines in the neighbourhood of the Cape of Good Hope , and if we compare this map with those of earlier epochs , we find the position of that line progressively more and more to the east as we ascend in the order of time .
Hence we should be led to expect that about this period it might be found to pass over the meridian of the Cape .
The observations which have been made daily at the magnetic observatory at the Cape , since its establishment in 1841 , give reason to believe that the westerly declination which had been increasing for above two centuries , attained its maximum in the year 1842 or 1843 .
In April 1841 the declination was 29 ?
05 ' west , in and April 1844 29 ?
06 ' west* .
The earliest observations at the Cape with which I am acquainted , are those of DAVIs in 1605 , and KEELING in 1609 .
( Purchase , Book iv .
ch .
6 . ? ?
1 .
and Book iii .
ch .
6 . ? ?
4 .
) According to these observations the declination in 1605 was 0 ? ?
301 east , and in 1609 0 ? ?
12 ' west+ .
The line of no declination probably therefore passed over the Cape about the year 1607 , and in 235 years the westerly declination has increased from 00 to 290 , ( omitting the odd minutes , ) or at an annual average rate of 7'V4 .
Observations at several intermediate epochs show that the progression of this change was at least not very far from being an uniform one .
If we divide the whole period into four equal parts , we should have Ot In the year 1607. .
00 In the year 1666 ... .
7 15 W. In the year 1725 ... .
14 30 W. In the year 1784 ... .
21 45 W. In the year 1843 ... .
29 0 W. In the appendix of HANSTEEN'S Magnetismus der Erde , p. 24 , we have the record of actual observations as follows : 0I In the year 1667 ... .
715 W. 16 27 W. In the year 1724. .
In the year 1780 ... .
22 16 W. We may therefore conclude that the westerly declination at the Cape , which for above 200 years had increased at an average rate of about 7'4 a year , or a degree in about eight years , has been for the last three years nearly stationary , having arrived at a maximum of 29 ?
and a few minutes about the year 1843 ; and that a decreasing progression may now be expected+ .
Ships passing the Cape , on a voyage to the * The observations at the magnetic observatory at the Cape of Good Hope , preparing for the press , will show the mean declination in each month of the years referred to .
t See also , for the latter observation , HANSTEEN'S Magnet .
der Erde .
Anhang .
S. 146 .
+ Captain FITZROY observed 280 30 ' in 1836 ; at that epoch , consequently , the maximum had not been reached .
Sir EDWARD BELCHER , in 1842 , observed 29 ? ?
13t .
, east , will find that the westerly variation , which increases the whole way from the Brazils to about the meridian of the Cape , begins there to diminish , and continues to diminish , passing into easterly variation increasing , for above 100 degrees of longitude east of the Cape .
The separating line which now passes through the Cape divides spaces distinguished by opposite magnetic characteristics ; on the west side of the Cape the north end of the needle moves to the west , and on the east side to the east , as east longitude increases .
The maps which exhibit the results of the observations in the two ships , of the Declination , Inclination and Intensity , in the voyage of 1841-1842 , and the isogonic , isoclinal , and isodynamic lines traced approximately in conformity with them , are a continuation orf the maps published in No. V. , which embodied in a similar manner the results of the preceding voyage .
The results in the Erebus are distinguished from those in the Terror by a different character , for the purpose of peirmitting the degree of accordance in the two series of independent determinations to be readily judged of by the eye .
These aps aord the best reply t se soto those who have expressed doubts of the success of observations of the inclination and intensity made at sea .
Magnetic lines , drawn from observations made in parts of the globe to which observation had not previously extended , are the proper test by which we may judge of the degree of approximation with which the values of the numerical elements have been obtained in a general mathematical theory of terrestrial magnetism , such as M. GAUss 's .
The portion of the observations of the Antarctic Expedition which has been placed before the Royal Society in No. V. and in the present number of these Contributions , permits us already to form some conclusion on this point .
Plate XIII .
exhibits the lines of one of the magnetic elements , i. e. the intensity , computed by M. GAUSS'S theory , and drawn in Plates XVIII .
and XIX .
of the Atlas des Erdmagnetismus , compared with the lines which are the direct results of observation .
The very imperfect resernblance between the two systems of lines is of course no impeachment of the sufficiency of the theory , with cortected numerical elements , to represent the natural phenomena in parts of the globe which observation may not have reached .
r'Ihe degree of approximation to which it will do this must depend upon the extent and correctness of the observation-basis from whence the numerical elements are derived , and upon the order of the magnitudes comprehended in the calculation .
The evidence which the plate affords , that the calculations in the elaborate work referred to differ so widely from the facts in the southern latitudes , shows how much observations were wanting in those latitudes for the purpose of perfecting the theory ; and is an ample justification ( if indeed any justification were necessary ) of the exertions which the last few years have witnessed to obtain them .
Since these pages were written I have received from Mr. ARCHIBALD SMITH the following note .
Regarding it as a continuation of the memrorandum with which he 116* was so obliging as to favour me , printed in the last number of these Contributions , I avail myself of this opportunity of giving it an early circulation .
" The apparent changes in the values of the constants a , b , c and d , in the Erebus and Terror ( Contributions , No. V. , p. 153 ) , seem to show that those vessels had an appreciable quantity of magnetism , which was so far permanent , as to retain for a considerable time traces of the inductive force to which they had been exposed , and perhaps some strictly permanent magnetism .
It seems , therefore , desirable to introduce into the expressions in the memorandum printed at p. 147 of Contribution No. V. , terms which will represent such forces .
" Suppose , then , as in the memorandum , that &lt; represents the total magnetic force of the earth at the place of observation , 0 the inclination , 4 the azimuth of the ship 's head , reckoning from N. to W. , and that p ' , 8 ' , ' represent the values of the same quantities shown by an instrument at a fixed position in the vessel , and affected by the attraction of the iron in the vessel ; and let P , Q , R represent the attraction of the permanent magnetism in the vessel to the bow , to the starboard side , and ver- .
tically downwards .
The fundamental equations of the former memorandum become by the introduction of these terms , p ' cos 0 ' cos ` =-P [ A ' cos 0 os B cos 0 sin +C sin 0 ] +P ' cos 6 sin -'=p [ D cos 6 cos +E ' cos 0 sin +F sin 6]\ +Q ' sin 0'=-p [ G cos 0 cos +H cos d sin ~+ K ' sin 0 ] + R. ' In these equations A ' , B , C , D , E ' , F , G , H and K ' are constants depending on the distribution of the soft iron in the ship , and perhaps on the temperature and other circumstances .
c " If we suppose , as before , that the soft iron is syrmmetrically disposed , the equations ( 1 .
) ( 2 .
) and ( 3 .
) of the former memorandum become , cos 0 ' cos 9l P A ' cos =cos +a tan + A ' Ccos ( 1 .
) cos 0e sin. Q A ' cos =b sin ; + A ' cos. .
*.(2 .
) ?
r sin Of RA cos 0 =ccos d tan +A cos c 0(3 .
) Let H represent the horizontal force =p cos 0 , H ' the affected horizontal force PQR= p'cos d ' , and let a tan d+ A-T=L , A , H = M , and d tan + AH = N. The last equations become HI XH cos = cos+L ... . .
( l )AH sin =b sin + M.(2a .
) HLtan T. An , H= ccos+ N ... . .
( 3a .
) 117* By the introduction of the same quantities , the equations numbered from ( 4 .
) to ( 14 .
) in the former memorandum become IH , -H = cos ' cos ' 4+ b sin 'sin C+L cos ( +M sin C ( cos + L ) sin = ( b sin + M ) cos I ; ... ... and representing ' , or the deviation , by , sin =I sin '-Mcos ' ( 1-b)sin cos C. *+ ^L sin -+ M cos +l+b sin ( + ' ) *b sin +M tan b sin co+ L ... ... ... .
' .
tan &gt; = cos^+L ccos ' +N= ( b sin m'+M ) cosec tan ... .
= ( cos + L ) sec ' tan ' ... . .
=-/ ( cos L)2 + ( b sin + M)2 .
tan .
cos +ZN c cosc-N tan =in sin + bM cos 4++N =c cos+L ... COS Co c cos r+N / ( cos + L)2 + ( b sin + M)2 " Equation ( 7 .
) may also be put under the form. .
( 4 .
) ... ... ( 6 .
) ... .
( 7 .
) ? . .
( s. ) A. .
**I ( 9.).* .
( 10 .
) ... ( 11 .
) ... ... . .
( 12 .
) ... ... . .
( 13 .
) . ?
... . .
( 14 .
) sin 8 -= b/ / L2+ M2sin ( ' - ) ++ sin ( 1+ ) , 2 1-bsecsin 2 i+ L sec p sin ( ' , ) +I+ sin ( ) , where M in which tanp , and p represents the displacement of the line of no deviation towards the starboard side .
" ; By means of these equations we can determnine A ' , L , b , M , c , N , from observations made at sea alone .
The first four of these quantities furnish the corrections for the horizontal force and the declination .
There is greater difficulty in obtaining the correction for the inclination .
It will be observed that 0 only occurs in these equations involved in the quantities L and N. If there were no permanent magnetism in the vessel , it would be necessary , in order to determine the correcting factors a and d , that observations of the inclination on shore , and corresponding observations on board , should be made in at least one magnetic latitude .
If there is any appreciable permanent magnetism , observations of the inclination on shore and on board , and of the horizontal force , should be made in at least two magnetic latitudes .
This would be sufficient if a , P , d , R remained absolutely constant .
As that appears not to be 118* the case , as many observations as possible should be made of the inclination on shore and on board , with corresponding observations of the horizontal force , Such observations should be made with great care when the vessel is on or near the magnetic equator and before and after any rapid change of magnetic latitude , and whenever the vessel returns to a place where the observations have been made before on board the same vessel , under the same circumstances as to the distribution of her iron .
f " When the permanent magnetism is symmetrically distributed , Q =0 and M 0 , and the other constants may be easily , and probably with great accuracy , determined from the following equations .
The small letter suffixed to the symbol of a function indicating the affected value observed with the vessel 's head on the N. , W. , S. , E. ( affected ) points , A't n ... ... .
( 15 .
) 2IH a tan 0+ -tL = H , H , .
( 16 .
) _ Hw + He.('7 .
) b= H ... ( 17 .
) 2 ?
Hn , n.H ... .
H , tan 0+ H tan 0d tan 0 A ' N Hn N H , 1 Hn tan , H , tan 0 , ?
H..-..(19 .
) " The values of HJ , H , , He , Hw , are given by the square of the number of vibrations of a horizontal needle made in a given time , and beginning to vibrate in a given arc , and require no correction except for temperature .
If n , s , represent the number of vibrations made by such a needle in the same time , with the ship 's head successively on the north and south points , and if A represent the value of 6 when = 90 , the values of L and A are given by the following simple expressions : If tan Xc L=cos2X ... ... ... .
( 20 .
) A=90 ?
-2 ... ... ... ... ( 21 .
) The equations ( 18 .
) and ( 19 .
) may be put under the form Rn sin 0 .
+p sin 0 , dtanO +R=N I_ 1f ISl+'S &amp; * l* ( 22 .
) dtan A'-H , = N= cos0t+f8coss ( 22.).n .
si n -f sin 6cf CO ?
8+ cos * ( 23 .
) and the values of N and c obtained , but probably with less accuracy , from obsenrvations of the total intensity and inclinations made with a Fox 's instrument .
" Note.-The last equation in the former memorandum is erroneous .
The value of , P cannot be obtained from two observations of the true azimuth of the ship 's head , when [ ' += 180 , independently of a. " 119

