]\gt ; III .
THE BAKERIAN LECTURE .
Magnetic Survey of the British Isles forth Epoch .
A. W. RUCKHIR , , F.R.S. , and T. E. THORPE , B.Sc. , Ph. D. , F.R. Lecture delivered April .
received June 6 , 1889 .
CONTENTS .
Page .
Introduction\mdash ; References to ormer Surve .
is . . . . . . .
55 Defects of Former Suryeys . . . . . . . . . .
55 Scope of the Snrvey ; Choice of Stations . . . . . . . . .
56 Epoch . . . . . . . . .
56 Instruments used . . . . . . . . . .
56 Kew as Base Station . . . . . . .
57 Comparison of Kew and Survey Instruments . . . . .
58 , 59 Units employed . . . . . . . . . . .
59 Comparison of Survey Instruments under Normal Conditions in . . . . .
60 Chronometers and Method of Determining Rates . . . . .
61 Number of Stations ; Average distance apart . . . . . . . . .
62 Method of Taking the Observations ( Adjustment of Mirror , \amp ; c. ) . . . . .
63 , 64 Discussion of Diurnal VariatioIt and Disturbance . . . . .
64-69 Method of Tabulating the Observations . . . .
69-76 Meaning of Symbols used . . . . .
70 Example of Declination Form . . . . . .
71 Horizontal Force : Independent ResuIts . . . . .
72 , 73 Specimen Form . . . . .
74 Inclination . . . . .
76 Errors of Experiment . . . . . . . . . . . . .
76-80 Declination . . . . . . . . . .
78 Horizontal For . . . . . . . . . .
78 Inclination . . . . . . . . . .
79 , 80 Secular Corrections . . . . . . . . . . . . . . . . . .
81-93 Dip . . . . . . . . . . .
81-87 In Declination . . . . . . . . . . . . . .
87-91 In Horizontal Force . . . . . .
, . . .
91-93 Form of Presenting ResuIts of Observations . . . . . . . . . . . . .
93-95 Description of , and ResuIts at , each Tndividual Station . . . . . . . . .
95-232 Calculation of the Lines:\mdash ; The Isogonal Lines .
233-244 Terrestrial and True Isomagnetics . . . . . .
233 Three Classes of Disturbances . . . . . . .
233 5.7.90 Page .
Page .
Method of obtaining Eqnations to Isomagnetic Lines . . .
234 , 235 Objections to Method of Least SquareR . . . . . . . .
236 ' ' Central Stations\ldquo ; . . . . . . . . . . .
237 Declinations calculated from the District Lines for Points Uniformly Distrib.uted over the Area of bhe Survey ( Table ) 238 , 289 Formula for calculating Declinations . . . . . . . . .
240 Discussion of lsogonals on Borders of English and French Surveys . . . . .
241-244 The Isoclinal Lines . . . . . . . . . . . . .
244-246 FormuJa for ting ditto . . . .
245 The Isodynamic Lines . . . . . . . . .
247-250 Nccessity for Two Formulae . . . . . . . .
243 Formulae for calculating Horizontal Force . . . . .
250 Summary of Elements and Disturbances . . . . . . . .
251-258 Scotland . .
, . .
251 , 252 England and Wales . . . .
253-256 lreland . . . . .
267 , 258 Supplementary Stations . . .
258 Discussion of Local and Regiona,1 Disturbances:\mdash ; Methods of deducing Disturbamces . . . . . . .
259-269 Eifect of Disturbing Centre on Isogonals . . . . .
261 , 262 other Lines . . . . .
263 , 264 , " " Ridge\ldquo ; and\ldquo ; Vall ' . . . . .
265 Nfeaning of iCal Force Disturbances Doubtful . . . .
266 ResuIts at Different Stations at Different Times . . . .
267--269 Table of Disturbing Forces . . . .
270-273 Surveys of Selected Districts . . . . .
274-292 The Malvern Hills . . . . .
274-276 The Island of Canna . . . . .
276-279 The E. and S.E. Counties . . . . .
280-286 The Wash . . . . .
286-288 The Leicestershire District . . . . . .
288-292 General Results of Investigation of Disturbances . . . . .
292-314 Methods of Testing Conclusions . . . . .
292 Comparison with Mr. WELSH'S Survey . . .
293 Declination at Stations common to the veys of 1857 and 1886 . . . . . . . . . . . . . . .
294-296 Disturbing orces in Scotland ( WELSH ) . . . . .
297-299 The Highland District . . . . . . . . . . .
299-302 The Scotch Coalfield District . . . . . . . .
302 , 303 N.W. , N. Wales , and Galloway . . . . . .
England . . . . .
306-308 The Southern District . . . . . .
308 , 309 Ireland . . . . . . . .
309 , 310 Magnetic Map of United Kingdom . . . . .
310 Relation between Magnetic and Geological Constitutions of Districts . . . . . . . . . . .
310-314 The Causes of Local and Regional Magnetic Forces . . . . . . . . .
314-320 Dr. NAUMANN'S Views . . . . . . . . .
314 Earth-currents . . . . . . . . . . . . . .
315 Tnsufficiency of observed to Produce ifects 316-318 Insufficiency of Geological FauIts . . . . . . . .
318 SURVEY OF THE BRITISH ISLES FOR THE EPOCH JANUARY 1 , 1886 .
55 Page .
Page .
Summary . . . . . . . . . . . . . . . . .
320 , 321 Tables for the Calculation of the Magnetic Elements . . . . . . . . .
322 Table VIII .
Declination .
IX .
Horizontal Force .
X. Dip .
Rate of Secular Change . . . . . . . . .
323-326 Table XII .
Rates of Secular Change . . . . . . . . . .
Two Magnetic urveys of the British Isles have been made previous to that of which an account is given in this paper .
The observations necessary for these were taken between the years 1834-38 and 1857-62 , and the results reduced to the epoch 1842-5 by Sir E. SABINE , in a paper published in the ' Philosophical Transactions ' for 1870 ( ' Phil. Trans vol. 160 , 1870 , p. 265 ) .
As a full account of both surveys is given in that paper , it is unnecessary to describe them in detail here .
The first was made by five observers , viz. , Sir E. SABINE , Captain J. C. Ross , Mr. R. W. Fox , and Professors LLOYD and JOHN PHILLIPS .
In the second survey ( 1857-58 ) , Mr. WELSH , Superintendent of the Kew Observatory , made an admirable series of observations in Scotland , though , unfortunately , the exposure to which he was subjected brought on an illness which terminated in his death .
Sir EDWARD SABINE made observations on the Force and Dip at 24 stations in England , and some declinations determined by several naval officers between the years 1855 and 1861 , were utilized .
Altogether observations were made at 243 stations .
( ' Phil. Trans vol. 162 , 1872 , p. 319 .
) It has , we believe , for some time been thought by those interested in terrestrial magnetism , that another survey of the United Kingdom should be undertaken , and we ourselves drew attention to the matter in a paper " " On the Irregularities in Magnetic Inclination on the West Coast of Scotland\ldquo ; ( ' Roy .
Soc. Proc vol. 36 , 1884 , p. 10 ) .
Not only was this desirable in order that the secular changes in the direction of the lines of equal Inclination , Force , and DecJination might be -determined , but also because the earlier surveys left much to be desired with regard to the distribution of the stations and the number of the Declination observations .
Thus the Declination was determined about the epochs 1836 and 1857 at 84 stations only , of which 11 were common to the two surveys .
The maps given by Sir E. SABINE in the paper already referred to show that different districts havo received very different degrees of attention .
Stations where the Dip has been determined cluster thickly about the coast of Scotland to south of Oban , about the English lakes , and the south coast of England , and are thinly distributed in the North of Scotland , in the eastern counties of England , and in the centre of Ireland .
In like manner , while ( owing chiefly to the labours of Mr. WELSH ) the Declination had been measured at 40 places in Scotland , it had been observed with adequate instruments at only 28 stations in England , and 16 in IrelaIld .
We have therefore undertaken , and , in the course of the five years 1884-88 , both inclusive , have completed a new magnetic survey of the British Isles .
In our opinion , it would probably have been better if a larger number of observers had been engaged in the task , so that it could have ) finished in a shorter time .
In fact , we originally made a proposal in this sense which was , we believe , brought before the Kew Committee of the Royal Society .
It was considered that the objections to the employment of many instruments and observers were sufficient to prevent the acceptance of such a scheme .
With proper organisation , we believe that the errors thus infxoduced would have been much less serious than those due to the uncertainty as to the true secular corrections at any particular station .
As , however , it was more important that the survey should be made than that it should be made under the best possible conditions , we have in undertaking the task ourselves devoted most of our vacations and spare time to it , so that there should be as little deIay as possible , and have collected all the facts which throw light on tlJe value of the secular corrections .
Although the re-determination of the lines of equal magnetic Declination , Force , and Dip has been the main object of our investigation , a number of other questions have naturally come under consideration .
We have made some alterations in the Kew magnetometer which have been described tu the Physical Society ( ' Phil. Mag August , 1888 , p. 122 ) ; the method of presenting the results of the experiments has been modified so as to afford a greater test of their accuracy ; the validity of the method of correcting for diurnal variation and disturbance , especially at places far diQtant from the base station , has been reconsidered , and finally , and perhaps chiefly , we have given more attention than our predecessors have done to the distribution and causes of " " local magnetic attraction In view of the difficulties caused by such disturbances , we have taken special pains to indicate the position of our stations as accurately as possible .
This has been done , not merely by verbal description , but by taking out the latitudes and]ongitndes from the inch Ordnance maps or the Admiralty charts , with far greater accuracy than is necessary for the calculations in which these quantities are afterwards employed .
For similar reasons we have selected , when possible , public parks , open commons , or other situations which are likely to be still available when the survey is repeated .
It must , however , be remarked that , at some places which were not included in our original programme , observations have been made mainly because a favourable opportunity presented itself , and that , in such cases , it was not always possible to exercise the same care in the selection of a site .
Epoch .
The epoch of the survey is January 1 , 1886 , to which date all the observations have been reduced .
Instruments .
survey of Scotland was mainly made with a set of instruments which belong to Professor RUCKER .
They are a Kew Magnetometer by ELLIOTT BROS .
, No. 60 , and a SURVEY OF THE BRITISH ISLES FOR THE EPOCH JANUA RY 1 , 1886 .
Dip Circle by Dover , No. 74 .
These instruments were also used by Professor RuCKER in his portion of the survevs of England , Wales , and Ireland .
The instruments employed by Dr. THORPE in these countries are the property of the Science and Art Department .
They are of the same patterns and by the same makers ; the Magnetometer and Dip Circle are numbered 61 and 83 respectively .
Arrangements have been made for placing this latter set of instruments in the Collection of tific Apparatus at South Kensington .
They will kept for surveying only , and will be available for comparison with instruments which may used by future observers .
Base Sta tion .
Our base station is the Kew Observatory .
Both sets of instruments were tesfed there before they came into our possession , and we have also made numerous comparisons with the Kew instruments during the of the survey .
We take this opportunity of expressing the great obligation we are under to Mr. .
M. WHIPPLE , the Superintendent , and to Mr. T. BAKER , the First Assist , ant in the Observatory , who , with the approval of the Kew Committee , rendered us every assistance in their power .
Our frequent visits and requests for information as to diurnal variation and disturbance have imposed much additional labour on these gentlemen .
The help we required has , however , invariably been given with a readiness and heartiness which merit our grateful We tested our instruments at Kew in 1884 , 1886 , and 1887 .
A very large magnetic disturbance set in on March 30 , 1886 , the 31st was also much disturbed and the effects were still to be distinguished on April 1 .
We have , therefore , neglected the observations made on that day , but , with tLis exception , the following tables contain all the determinations made by us in the Magnetic House .
The first three columns give the date , instrument , and observer , and that headed the value of the eJement obtained .
The numbers given are not corrected either for diurnal tion or disturbance .
In the column headed is the corresponding value read off irom the curves of the continuously recording instruments at Kew , standardised by means of the monthly observations which are taken there .
The differertces are giv.en in the next column .
They are very nearly stant , but are rather larger than we should priori have expected .
Our own instruments are evidently in accord .
Subtracting the mean difference from the Kew values , we should , if both sets of observations were perfect , reproduce our own numbers , so that a comparison ween columns and gives the error of experiment and comparison which is bulated in the last column .
SURVEY OF THE BRITISH ISLES FOR THE EPOCH JANUARY 1 , 1886 .
To obviate , as far as possible , the use of decimals , the Forces are throughout this paper expressed in terms of metric units ( metre , gram , second ) , and all numerical values of Forces can be reduced to C.G.S. units by dividing by 10 .
In the case of the Dips , of which no continuous record is kept , the following plan was adopted .
We give , under the column , our results corrected for diurnal variation and disturbance as hereafter described , each being the mean of the two needles .
In column are the values obtained a.t Kew at the nearest date to the time of our observations taken from the published record .
The dates in question are July 28 and 31 , 1884 , Sept. 22 , 24 , and October 25 .
26 , 1887 .
The mean of the last two is taken as applying to the whole period of the 1887 observations .
The used to form an image of the Sun which can be viewed through the telescope were tested independently .
We were , however , unfortunate in that on most of the days when we observed at Kew the Sun was invisible .
The idian mark also , which is used in the measurements , cannot be seen from the lawn imme ( iiately in front of the lsIagnetic House , and we were , therefore , obliged to use a rod which was fixed at some distance , as nearly as might be in the meridian line in which the instrument was also placed .
On one occasion ( October 11 , 1887 ) the latter adjustment was inaccurate , and a correction was applied deduced from the angle subtended at the pillar in the gnetic house by the centre of the tripod and the rod , and from the distances of the tripod aJld pillar from the rod .
At Parsonstown the instrument was placed in the meridian line used for the ervatory of the Earl of ROSSE , and a check on the declinometer was thus obtained .
From what has been said it will have been seen that the determinations of the geographical meridian at Kew were not made under the most favourable conditions , but the following Table is sufficient to show that no important error attached to the indicat , ions of the instruments .
AZIMUTH of Kew Meridian Mark from Sun Observations .
At Parsonstown the difference between the reading for the fixed mark and the me1idian , as calculated from a sun observation , was We also thought it would be desirable to compare the instruments under the normal conditions of use in the field .
We therefore made three sets of observations at Ard Point , Stornoway ; at Bunnahabhain , in the Sound of Islay ; and at Stranraer .
At the first station the results of the individual experiments made with each instrument were in good accord , but , as the declinations differed from each other by an amount very much greater than any possible error of experiment , viz. , 18 ' , it was obvious that the station was affected by local magnetism .
At Bunnahabhain we took the precaution to exchange positions ; these observations were made on a patch of limestone , and were in agreement .
At Stranraer the sky became overcast , and it was impossible to move the instruments after the first determination of the geographical meridian .
The results show that the station was a good one , and we therefore give here the declinations obtained at Bunnahabhain and Stranraer .
As we shall have to refer to them again , we reproduce only the data which bear upon the poiut under discussion .
At Bunnahabhain the thread of instrument 61 was broken , and had to be replaced in the field , which accounts for the fact that the first observation made with it is not in very good accord with the others .
The Greenwich mean time is reckoned throughout from midnight .
SURVEY OF THE BRITISH ISLES FOR THE EPOCH JANUARY 1 , 1886 .
STRANRAER.\mdash ; Aug. 28 , 1888 .
During the survey we employed various chronometers , some of which were hired from Messrs. DENT , and others were lent to us by the late Professor BALFOUR F.R.S. , and by Captain WHARTON , R.N. , F.R.S. , Chief Hydrographer to the Admiralty , to whom our thanks are due .
We enjoyed an important advantage over our predecessors in that we were able to determine the rates of our chronometers frequently by comparison with Greenwich , by means of the.10 .
and 1 signals , of which the former is sent .
to all post-offices in the domo .
We have to thank Mr. PREECE , F.R.S. , Chief Electrician to the General Post-office , and Mr. J. C. LAMB , the Head of the Telegraph Department , .
for their kindness in giving or obtaining for us permission to the signals .
of the local post-office officials not only afforded us every facility for correcting our chronometers in accordance with their instructions , but gave us additional help in the selection of suitable stations .
When possible , we received the signal every day , and rarely omitted more than two consecutive days .
At places where the signal had undergone one or more re-transmissions , an error on this account was inevitable .
By the kiudness of the Earl of ROSSE , we were able to determine its magnitude on the occasion of our visit to Parsonstown .
The time signal was observed both by Professor RUCXER and Herr BoDDEKKER , the Superintendent of ROSSE'S Observatory , who agreed exactly as to the error of the chronometer .
The value thus obtained differed by 4 sec. from that given by the observatory clock , the error of which was known from star observations .
We are inclined to think that this amount was rarely exceeded , or the es would have varied more widely than was actually the case .
No special precautions against error had been taken at Parsonstown , whereas in many districts we had , when practicable , made arrangements at the central office that particular care should be taken on those mornings when we informed the authorities that we intended to receive the signal .
We have on several occasions received at the same post-office signals sent by two different routes , involving re-transmission , and never detected any appreciable difference .
When visiting outlying stations at sea we kept one chronometer on board in a fixed position .
This served as the standard .
Another instrument , which was frequently compared with it , was used for making the comparisons with Greenwich and for the work of the actual observations .
At these periods of our survey longer intervals elapsed between successive receptions of the time signal , during which we depended on the standard .
With the exception of two or three cases , which are referred to in the detailed account of the observations at the stations affected , we do not think that any error exceeding a minute of arc can have been introduced into the declinations from uncertainty as to the true time .
of Stations .
In selecting stations we have aimed at uniformity of distribution over the whole area under investigation , and , subject to this condition , have chosen places at which observations have previously been made .
We .
have also avoided situations where disturbance by so-called local attraction was known to be great , except in special cases to be discussed hereafter .
On many tours we have carried geological maps which have aided us in choosing sites .
In all we have made observations at 54 stations in Scotland , 102 in England , Wales and the Channel Isles , and 44 in Ireland .
At many of these places which are counted as a single station our instruments have been set up in different localities in order to study or eliminate the local effects of rocks .
These sub-stations were in some instances only a few yards and in others several miles apart .
The more important of them are indicated by letters in the lists given hereafter , and if these are added they bring the number of stations up to 213 .
In addition to these we observed on the island of Canna at 23 places .
If we took minor changes of position into account this total would be considerably increased , and as we have been able to use the observations made at Greenwich and Stonyhurst , and the data for Cherbourg and Berck-sur-Mer , furnished by the survey of France , recently completed by M. MOUREAUX Determination des Elementl Magnetiques en France par M. TH .
MOUREAUX .
Paris : GAUTHIER-VILLARS , 1886 ) , our conclusions are based upon observations made at about 250 different places .
The average distance apart in normal districts is about 30 miles .
At most of the principal stations we made two Dip observations , one with each needle , and a complete set of measurements for the determination of the Declination and Horizontal Force .
At many places we determined the Declination twice , making two independent sets of observations on the Sun and the needle .
In some cases only the geographical meridian or only the direction of the gnetic axis of the needle was re-determined .
When time was short or the weather unfavourable the deflection experiment was omitted .
The following Table shows t , he total number of observations made .
By magnetic meridian we mean determination of the direction of the magnetic axis , which , when combined with the Sun observations or geographical meridian , gives the Dec]ination .
We do not include 23 Declinations taken on Canna by means of an azimuth compass and chart .
*Exolusive of five supplementary stations along the valley of the Wye , at which Dip observations only were made in 1889 ( vide p. 84 ) .
SURVEY OF BRITISH ISLES FOR THE EPOCH JANUARY 1 , Most of these observations were taken by one or other of us .
We have , however , to thank Mr. A. P. LAURIE , now Fellow of King 's College , , for observations of Dip at 8 stations .
The stations on the West Coast of nd were , for the most part , visited in Dr. " " Coventina During first.year ( 1884 ) we generally observed together .
Afterwards , in order to save time , we avelled separately .
of the Observations , The conditions and time of the observations neoessarily varied , but the greatest number were made as follows:\mdash ; Shortly before 10 .
we visited post-office to correct our chronometers by the tinle signal from Greenwich .
We then drove to the station which had been selected the night before or earlier in the morning .
The first observation taken was the solar azimuth , which was finished about an hour before noon .
This was followed by the declination , vibration , and deflection in this order .
The dip was then , and , if time and weather allowed , we often repeated the azimuth and declination .
The Sun observations were bardly ever taken within an hour of noon .
In a variable climate like that of the United Kingdom the weather often a serious difficulty .
We ied with us a small tent , in which the dip and vibration ations could be made , were also provided with waterproof covers for the Magnetcmeter and Dip Circle .
We were thus to make the Dip observations , for example , during showers , while the Magnetometer , though outside the tent , was protected by its covering .
The case containing the dip needles was carried in a filled with soda-lime to prevent } axles being injured by rust .
During sunshine we shaded the deflecting in the deflection experiment by cardboard case , or by throwing a light piece of cloth over it .
At 42 principal and subsidiary stations the force was determined by means of the vibration experiment only , the deflection omitted and the moment of the magnet deduced from values obtained at neighbouring stations .
The only part of the observations which requires special comment is the of the mirror employed in the sun observations .
The adjustment can be effected by means of the reflected image of the cross wires in the telescope .
We , however , that in actice in the field this method is troublesome .
It is difficult to see the image unless the mirror is very bright , and there is fairly strong sunshine .
On days when the weather and sunsbine are uncertain it is very annoying to have to spend time when the sun is visible in making preliminary adjustments .
From and after our visit to Stornoway in 1884 , i. e. , after observations had been made at about a dozen stations , we adopted a different plan .
The mirror was frequently adjusted either indoors in ccordance with the directions of the Admiralty 'Manual , ' or by means of some elevated object in the field .
In the latter case the error of parallelism was first got rid of by observing the objeot when the observer 's back was turned towards it , and the mirror was nearly vertical .
The mirorr was adjusted until the image appeared in the same position as before when it was reversed in its .
The object was then viewed by reflection when the instrument was turned through l80o , i.e. , when the mirror was nearly horizontal and the axis of the mirror was made perpendicular to the axis of the telescope .
By repeating these processes twice or thrice a perfect adjustment was obtained , and it was found that under ordinary circumstances the instrument could be carried about from place to place for some time without this adjustment being seriously affected , To make certain , however , that all was right , we always ( when possible ) took observations of the sun both in the " " front\ldquo ; and " " back\ldquo ; positions , i. e. , with the mirror approximately horizontal and vertical and reversed the mirror in its bearinae in both positions , By the latter precaution the error of parallelism ( if any existed ) was eliminated , and the observation thus afforded data for the calculation of the error of oollimation , and for the correotions which in consequence of this error must be applied to the " " front\ldquo ; observations .
By this method the actual adjustment of the mirror could be made at any time or .
If the time at our disposal was short , or the weather uncertain , the instrument was regarded as ready for use as soon as it was set up in the field , and thus no time was lost , but the observations were so taken that any error of adjustment could be calculated and allowed for .
Lastly the agreement of the measurements made in the front and back positions gives a valuable test both of the adjustments and observations .
In the case of the observations made in 1888 with Ma , gnetometer No. 60 , the method of correoting the mirror by the image of the cross wires was again resorted to .
Some modifications of the strument which we have introduced ( loc. cit. ) now make this method practicable , and it was found to work well under the new conditions .
In the survey of 1857 oorrections for diurnal variation and disturbance were applied to the declination observations only .
These two quantities are intimately connected .
The total deflections of the continuously recording instruments at any instant are the sums of the normal diurnal variations and the disturbances .
If the latter are assumed to occur simultaneously and in equal intensity over an area such as that of the United SURVEY OF THE BRITISH ISLES FOR THE EPOCH JANUARY 1 , 1886 .
Kingdom , and if the G.M.T. at which the observations were made is known they may be determined from the curves obtained at Kew .
On the other hand the magnitude of the diurnal variation at a given instant depends on the local time , and in the case of stations so far apart as Kew and Valentia , the diurnal variation of the declination would , between 10 and .
G.M.T. , differ by several minutes of arc .
We have therefore taken out from the Kew curves the deviations from the mean at the G.M.T. of our observations , and subtracting the corresponding value of the diurnal variations at Kew , have called these differences the temporary disturbances at the time of the observations .
We have assumed further , that the Kew mean curves of diurnal variations may be taken as applying to the whole of the United Kingdom , but we have , in cases , used the correction corresponding to the local time .
Thus the total correction is the algebraic sum of the diurnal variation at the local time , and of the disturbance registered at Kew at the G.M.T. at which the observations were taken .
As proof of the ] idity of this method , we have picked out all the stations in Ireland at which two or more observations were made at times ) the diurnal variations differed by more than four minutes .
As full particulars are given later on , it is only necessary here to tabulate the corrections for diurnal variation and disturbance in addition to the final results .
It may , however , be remarked that in the majority of cases the determinations of the geographical , as well as of the magnetic meridian , are quite independent , and that at Athlone , Gort , and Westport , the results given were obtained on rent days .
The letters and are used throughout this memoir to indicate the departure from the mean value of the element at the time of experiment , due to variation and to temporary disturbance respectively .
The following are the best stations for the application of a similar test to Scotland .
The two sets of observations at Scarnish and Portree were taken on two different days .
Those at Stornoway were on the same day , but at different stations .
The observations at Bunnahabhain and Stranraer , which have been already discussed , are here arranged in order of time without reference to the station or instrument:\mdash ; In some cases where there is an appearance of a regular change , such as would be caused if the diurnal variation was not properly estimated , it can be otherwise accounted for .
Thus the first four observations at Bunnahabhain were independent in evely particular , but the last two depend upon the same Sun observation as the urth .
This probably accounts for the somewhat closer agreement in these cases .
The magnitude of the errors which would be introduced if the disturbance corrections were omitted can be estimated by noting that at Cavan the first and last observations when uncorrected give and , those at Portree and and those at Scarnish and respectively , instead of the much more accordant numbers entered in the Table .
These results when corrected are quite as good as those we obtained when observing at Kew , and furnish a complete posteriori justification of the method of correction .
The values of the diurnal variation employed for the year 1884-5 were furmshed to us by Mr. WHIPPLE .
For and 1886-7 we used the Tables given in the reports of the Kew Committee for those years ( ' Roy .
Soc. Proc vol. 41 , p. 415 , and vol. 43 , p. 226 ) .
These differed rather markedly as to the magnitude of the maximum variation from the curve for 1884-5 , the difference amounting to about .
As the report SURVEY OF THE BRITISH ISLES FOR THE EPOCH JANUARY 1 , 1886 .
of the Kew Committee for 1887-8 was not published when the observations made in 1888 were reduced , we used the same Table as for the 1887 observations .
The question as to whether we should apply similar corrections to the other elements has been carefully considered , and we have on the whole decided to do so .
Though we have worked up the observations on the Horizontal Force so that each experiment gives two nearly independent results , the times at which these have been taken are practically identical , and they do not serve as a test of the advantage of applying corrections .
In 18 cases we have repeated either the deflection or the vibration experiment ( generally the latter ) in the field on the same day .
The differences between the results when uncorrected and corrected for diurnal variations and disturbance are as follows .
They are expressed in terms of metric or C.G.S. The values of obtained by us at Kew in 1887 also afford a test of the method of correction .
The diurnal variation and disturbance are expressed in the following Table in terms of metric unit : llIl .
A. W. RUCKER AND DR. T. THORPE ON A MAGNETIC As then the diurnal variation between the hours at which our observations were generally taken amounts to metric unit , and as the probable difference between the corrected semi-independent results at the statiorJs referred to in Table I. , p. 67 , is 0004 ( the probable error being half that quantity ) , and as the probable error of the seven Kew observations is less than , there seems no doubt that the correction ought to be applied .
In the case of the Dip the diurnal range ( which is more than 2 ' ) exceeds the error of experiment , but the application of a correction for disturbance is more difficult and uncertain .
It has to be deduced from the Vertical and Horizontal Force magnetograms and thus the chances of error are increased .
The quantities to be dealt with are generally small , and the uncertainty of the readings taken from the curves unquestionably affects the results .
We have not many data which furnish any posteriori arguments of importance , as in most cases the times at which the two Dip observations were taken were separated by too short an interval for the diurnal variation to have altered much .
The observations made at Kew in 1887 are , however , decidedly improved by correction .
SURVEY OF BRITISH ISLES FOR EPOCH JANUARY 1 , 1880 .
On the whole , then , we think the application of the correction for diurnal range advantageous , but the correction for disturbance appears to be somewhat uncertain in its effect as applied to the field observations , and we have retained it chiefly for the sake of with our treatinent of the other elements .
The formulae used in the application of the corrections to the Horizontal Force and were as follows : \mdash ; If and be the values of the Force and the moment of the magnet deduced from the uncorrected and corrected observations respectively , aod if and be the total increments of the Horizontal Force due to diurnal variation and disturbance at the times when the deflection and vibration experiments are made , ; As the disturbances of the Dips have to be deduced from those of the Horizontal and Vertical Forces , the calculations involved are rather troublesome .
If be the change , expressed in minutes of arc , produced by increments and in the Vertical and Horizontal Forces respectively , where and are quantities which depend on the Dip and Horizon Force at the station .
Tables were prepared in which the values of these were entered for each complete degree between and , and for each tenth of a unit of Horizontal Force between and .
As they vary slowly , the value corresponding to any Dip or Force could be readily determined , and the value of was thus deduced .
Method of Tabulating the Observations .
We have ventured to modify considerably the methods ordinarily adopted of senting the results of a survey such as our own .
It is often very difficult for any one who studies the records of magnetic observations to learn anything as to whether the instruments were in good or bad adjustment , or whether the actual observations were careful or careless .
Sometimes superfluous data are given which supply no information on these points .
Thus , the quantities and , or and which are sometimes tabulated in records of Force observations , add but little to their .
In an observatory , indeed , they ought to be nearly constant from month to month ; but as , in both of them , the changes in the value of the moment of the magnet and those in the Force are mixed up with the error of experiment , it is not easy to draw any definite conclusion from them .
To test the observations , we should compare quantities the difference between which ought theoretically to vanish .
In the case of a survey , and must vary largely from station to station , and they are , therefore , quite worthless as tests of the observations .
There are now five British magnetic observatories working with instruments of substantially the same patterns , and we think it would be very desirable if an agreement could be come to as to the form in which the results of the observations are published .
In own case we have been anxious to check the numbers given by experiment in every way , as it is only thus that we are able to determine what reliance may be placed on observations at stations where time or weather prevented the repetition of one or more of the measurements .
The following is an example of the method of tabulation adopted in the case of the Declinations :\mdash ; 2 is the interval between the southing of the Sun and the mean of the times at which the " " front\ldquo ; observations were made on the Sun 's azimuth .
It is taken as positive if the observations were made after noon .
Att. signifies the altitude of the Sun above the horizon at the same tiin is the correction made by means of the " " back\ldquo ; observations to the geographical meridian determined lrom the front observations alone .
This quantity serves to indicate the order of the error that may have been introduced by the omission of the back readings on some occasions .
G.M. is the reading on the horizontal circle which corresponds to the geographical meridian .
In most cases two determinations of the geographical and magnetic meridians were made at an interval of some hours .
In this time the tripod may have shifted slightly , owing to unequal sinking of the ground , warping of the legs , \amp ; c. Hence , if no fixed mark was read , each geographical meridian must be regarded as corresponding only to the magnetic meridian which was observed ] at the same time as itself As it is often difficult to select a suitable fixed mark , we have , whether a fixed mark was observed or not , regarded each reading for the geographical meridian as corresponding to that for the magnetic meridian which was nearest to it in point of time .
In the case of the magnetic meridian ( the reading for which is indicated by M.M. ) we tabulate the following data : ( 1 ) The G.M.T. at which the observation was made reckoned from midnight .
( 2 ) Half the difference between the readings when the magnet is erect and inverted is the angle between the magnetic and geometrical axes of the magnet .
If then from this quantity we subtract its mean value as determined from all the observations made in the year , we obtain a measure of the error due to inaccuracy of reading or adjustment , and to imperfect compensation of the diurnal variation or of any disturbance which may have been occurring at the time .
( 8 ) is the error which would have been caused had the torsion of the thread not been allowed for .
( 4 ) and are the deviations of the element from its mean value due to diurnal variation and disturbance respectively .
The first , as has been already explained , is calculated for local time from the Kew curves of diurnal variation , the latter is the SURVEY OF THE BRTTISH ISLES FOR THE EPOCH JANUARY 1 , 1886 .
disturbance registered by the Kew curves at the same G.M.T. as that at which the observations were taken .
( 5 ) 8 ( obs .
) is the value of the Declination obtained by subtracting the reading for the magnetic from that for the corresponding geographical meridian .
is the secular correction to Jan. 1 , 1886 , obtained as is hereafter explained .
( 7 ) 8 ( red .
) is the mean value of the Declination reduced to the epoch of the survey .
As an example we take Athlone , where observations were made on two days .
On May 8 , no back observations were taken , so that could not be calculated .
DECLINATION .
Athlone .
Observer , T. E. T. Magnetometer 61 .
In the case of the Horizontal Force observations the deflections were always taken when the distances between the centres of the magnets were and metre .
If then and are the common logarithms of the values of obtained at the two distances , the factor used in the corrections may be calculated by the formula of which the second term is very small .
( ' ture , ' Aug. 13 , 1887 , p. 366 ; Sept. 29 , ; Dec. 8 , p. 127 ; and Jan. 19 , 1888 , p. 272 ) .
The following table gives the mean values of obtained for each year during the survey .
The ordinary method of determining the Inclination with two needles gives two independent values , the agreement of which furnishes a test of the accuracy of the observations .
It is , however , usual to give only one value of the Horizontal Force deduced from the means of the values of and calculated from the deflection and vibration experiments .
The constancy or regular change of the value affords a means of determining whether the observations are sufficiently good , but it seems to us that nothing is for this purpose so satisfactory as the agreement or disagreement between two independent experiments .
The results of the ordinary deflection and vibration observations can readily be presented in such a form as to satisfy this condition with only an insignificant addition to the labour of reduction .
The vibration experiment furnishes twelve independent determinations of the time of 100 vibrations , six of which are taken when the scale of the vibrating magnet is apparently moving to the right , and six when it appears to be moving to the left .
If we take the mean of the six observations furnished by the first and last three of each of these groups , we have two independent determinations of the vibration period , based in each case on six observations , of which three were taken when the movement appeared to be to the right , and three when it appeared to the let If we agree always to combine the first and second of these groups with the values of obtained when the distaIlce between the magnets is and metre respectively , we obtain from each experiment two values of and H. Except in so far as errors may arise in the determination of the temperature , or in the adjustment of the instrument ( levelling , \amp ; c these are absolutely independent , and thus furnish a satisfactory check on the accuracy of the observations .
It would , however , be hardly worth while for the sake of this advantage to go through the labour of repeating the reduction of the vibrations .
As the two vibration periods are very nearly the same , this is unnecessary .
Let be the time deduced from all the observations , and the times given by the first and second groups of observations selected as above described .
Then if be the value of given by the deflection experiment , SURVEY OF THE BRITISH ISLES FOR THE EPOCH JANUARY 1 , 1886 .
and in like manner Thus , if we use the of the times for deducing two values of and or from the two deflection experiments ( as is done in the ordinary method of reduction ) we may obtain the independent values of and by altering the means in the ratio of the independent to the mean times .
Thus , at Horsham , the mean time of vibration was , and using this the deflections at the two distances gave and and and respectively .
The actual periods of vibration for 100 cillations observed were\mdash ; Taking the means of the observations recorded above and below the line we get for the time of one oscillation and These independent times differ from the mean by and per cent. respectively .
Hence , altering the values of and in the same proportion , but in the opposite direction to that indicated by the signs of these quantities , we get and and and fDCCCXC .
These quantities are quite independent , except as regards the temperature corrections and instrumental adjustments , and their difference is a ) sure of the rors of vation and imperfect compensation of the variation and disturbance , In tabul a the Force ions we give the G.M.T. at which the deflection and vibration vations were made , and the extreme temperatures during the .
progress of the the observed deflections and tho independent times calculated as above described .
The agreement of the times of vibration serves to test the vibration experiments .
As a measure of tho accuracy of the deflections we take the ra.tio of the two values of calculated by combining the mean time of vibration with each of the two deflections .
The ] rithms of these two quantities occur in the calculations , and all that is is to subtract them look out the corresponding natural number .
As quantity is very nearly unity we multiply by 100 and give the part only .
The symbol is thus the percentage difference between the results given by the deflections when combined with the mean time of vibration .
following form , which is which has been used , explains itself .
The results of the deflections diflered by per cent. , the times of by .
There was no disturbance , and the Horizontal Force , owing to diurnal variation , was and metric unit above the , at the times which } correspond to the of the deflection and vibration experiments respectively .
The corrected values are obtained from the mean values by applying the corrections for diurnal variation and disturbance .
FORCE .
Horsham .
Observer , .
W. R. Magnetometer 60 .
SURVEY OF THE BRITISH ISLES FOR THE EPOCH JANUARY 1 , 1886 .
At places where the deflection ' experiment was not made , the moment of the magnet was calculated from the other moments observed during the same tour .
It was assumed that the strength of the magnet diminished regularly , a linear expression was determined by equations of condition which gave the rate of decrease .
A careful analysis of the Scotch observations , which were the first reduced , proved that this method was satisfactory .
The greatest difference tween observation and calculation occurred at Kirkwall , and amounted to per cent. It is certain that in this case ) ) is not due to errors of experiment , to a real change in the moment , as the obseryations obtained at the next station ( Lerwick ) confirmed it .
We are aJso able to assign a probable cause for the iteration , as the " " Coventina\ldquo ; was caught in a " " roost\ldquo ; off Stromness , and experienced rather rough treatment , during which the magnet may have been jarred .
The mean difference ( irrespective of sign ) at all the stations was per cent. ; and if we put aside about one-fifth of the whole number of stations at which , to bad meteorological or magnetic weather , or some similar cause , the conditions under which the observations taken were not very favourable , the mean difference at the remainder was per cent. It must , of course , be remembered that this quantity is not a measure of the error of experiment , but of the algebraical sum of that error , and of the deviation of the rate of decrease of the moment of magnet from perfect nniforrnity .
The conclusion we came to from the discussion of the Scotch observations was thab if he rved moment differed from that calculated by the linear formula by per cent. , ) a real change had taken place in the nlagnet or the observation had ffected by some disturbing cause , which it was in general easy to specify .
In the later tours ths moments of the magnets changed very slowly , and it was sufficient to take the mean of the moments obtained at one or two stations immediately preceding and following that at which the deflection experiment had been omitted .
The total number of stations at which vibrations only were observed was 42 .
In the of the Dip observations the correction for diurnal variation was deduced from the Kew curves for the diurnal of the Horizontal and Vertical Forces , which were supplied to us by Mr. WHIPPLE .
The disturbance corrections were calculated as previously described .
The following form was used in working up the results:\mdash ; INCLINATION .
Bunnahabhain .
Observers : Needle 1 , A. W. R. ; Needle 2 , T. , T. Dip Circle , DOVER , 83 .
All the observations were tabulated in these forms , we propose to deposit copies of them with the Royal Society .
In this paper we give fewer data .
It will , however , be more convenient to describe these under the heading \ldquo ; Results of the Observations p. 93 , and we now proceed to use the tabulated results in a discussion of the accuracy of our work .
Errors of Experiment .
The completion of so large a number of observations as those involved in our survey , carried out for the most part by two observers with different sets of instruments , affords a good opportunity of testing the accuracy of experiment in the field , especially as we have followed a regular practice of determining both the geographical and magnetic meridians by means of independent experiments made at an interval of several hours .
Again , our method of tabulating the results of the Force observations furnishes a good test of the various parts of the experiments , and , as will be seen hereafter , we are able , in the case of the Dip observations , to arrive at conclusions as to the small errors due to the imperfections even of the excellent needles supplied by Mr. DOVER .
SURVEY OF THE BRITISH ISLES FOR THE EPOCH JANUARY 1 , 1886 .
In the first place then , we take as the error of a Declination , the difference from the mean of the values obtained at the same station at an interval so small that the secular correction is not involved ( i.e. , a few hours or days ) , when both the geographical and magnetic meridians are determined The error of a Force observation is taken as half the difference between the two independent values , calculated as above described .
Lastly , the error of a Dip observation is half the difference between the results given with needles 1 and 2 .
In the case of the last two , it is evident that both quantities may be affected with small errors which are thus detected , such as that due to uncertainty as to temperature , in the case of the Forces , and errors of setting in the magnetic meridian , in that of the Dips .
Taking , however , these quantities and treating them as though all the results in each group were measures of the same quantity , we get the following values of the probable error:\mdash ; In this Table , we count each double observation as one .
The fraction , in the case of the Declination , is due to the fact that , at some stations , an odd number of observations was made .
It is useful to analyse the observations still further .
In so doin , we have generally treated Scotland as a whole .
During our earlier tours we observed together , and sometimes one observer would take the geographical and the other the magnetic meridian , one would observe the deflection and the other the time of vibration .
It is , therefore , difficult to separate the results .
In England and Ireland we always observed apart ; and , with the exception of a short tour in 1886 , when Dr. THORPE used Dip Circle No. 74 , we always used different instruments .
The mean values of the quantities are taken irrespective of sign , so that they indicate the average value whether positive or negative .
The definitions of the quantities used as tests , are given on .
The symbol indicates the mean error of the element .
DECLINATION Observations .
Roughly then , the average errors due to imperfect adjustment of the mirror , to inaccuracy in determining the magnetic axis of the magnet , and to torsion , ought not , in each case , exceed about of a minute of arc , that no correction was attempted , as would be the case , for instance , with regard to , if no back observa ions of the sun were taken .
As in the great majority of cases , all three were corrected by front and back observations on the sun , by the magnet erect and inverted , and by determining the torsion , we may be sure that the actual errors were much below these amounts .
It seems , therefore , probable that the observations are affected with small errors which are not so capable of correction , for the mean and probable values of appear to prove that it is an even chance that each of two independent declinations differ from their mean by more than In the following we give a similar analysis of the Force observations .
The quantity is the percentage difference of either deflection experiment from their mean .
is the average difference in seconds between each " " independent\ldquo ; determination of the time of a oscillation ( as defined on p. 73 ) and their mean .
is the average difference between each independent determination of the Horizontal Forces and the mean of the two involved in each complete experiment .
In all cases the ages are , of course , taken without reference to HORIZONTAL Force .
SURVEY OF THE BRITISH ISLES .
FOR THE EPOCH JANUARY 1 , 1886 .
We have analysed the dips in a somewhat different way .
If one needle tends to give a slightly higher or lower result than the other , the mean value of the difference of the results , always taken in the me order\mdash ; viz. , dip given by Needle 1 , minus dip given by Needle 2\mdash ; ought to give the bias or measure oi the difference due to imperfections in their construction .
If this quantity be subtracted from the difference of the results in each observation : we the error of experiment , which is the difference due to imperfection in .
and correcting .
The mean of the last quantities , taken without reference to sign , is the mean error of experiment .
Thus , if we take the difference as positive when Needle 1 is the higher , out of 23 stations in Ireland at which observations were made with the same instrument ( 74 ) , it was positive at 15 and negative at 8 .
The mean of all the differences showed that , on the average , Needle 1 gave results higher , and Needle 2 results lower , than the mean of the two needles ; so that the re is evidence of a which would make Needle 1 give results higher than Needle 2 .
That this number is wort , hy of credit is shown by the close agreement of the values obtained with the same instrument in Engl : Jnd , and by the practical identity of all the obtained for Dip Circle 83 in England , Scotland , and Ireland .
The axles of the needles of Dip Circle No. 74 appear to have undergone some slight alteration between the end of and the beginning of 1886 , there seems no doubt that since that date Needles No. 1 in the two instruments have read about and respectively above the 1nean of the two needles .
The following Table gives a summary the results:\mdash ; INSTRUMENT 74 .
INSTRUMENT 83 .
An analysis of the errors and COl.rections such as we have carried out appears to constitute a very delicate test of the observer and instrument combined .
Thus it is curious to note that ( although the differences are small ) one of us has on the whole been more successful with the Declinations and the other with the Force observations .
As has already been remarked , the observations in Scotland were often divided between us , so that we do not attempt to analyse them , but both in England and Ireland Dr. THORPE'S Declinations agree the better by about , and Professor RUCXER'S Forces by about metric unit .
The facts that in Ireland Professor RUCKER'S mirror appears to have been generally in the better adjustment , and that in England Dr. THORPE'S observations of the period of oscillation of the needle were the better , indicate that the causes of the differences in the results were probably a more perfect management of the torsion of the thread in the Declination observations , and a slight superiority in the deflection observations in the Force measurements .
It will , we t.hink , be granted that in all cases the observations are as good as the occasion requires , and we discuss these minute differences not because we think they produce any appreciable effect on the results of our survey , but as an illustration of the fact that our method of tabulating the observations , conveys information as to the personal equations of the observers which is not afforded by the results as ordinarily presented .
We think it possible that if we had been able to carry out the work in a more leisurely manner , to observe azimuths only when the Sun was low in the heavens , and to wait for fine weather , the results might have been in closer accord , but on the other hand we do not think that any improvement so obtained would have been of practical importance .
The uncertainty of the value of the secular correction at any given station , and the changes produced disturbed districts by a slight alteration in the position of the instruments , are far more important than any residual errors by which our observations may possibly be affected .
OF THE BR , ITISH ISLES FOR THE EPOCH JANUARY 1 , 1886 .
The secular correction is often appreciably different at neighbouring stations .
Thus the annual change in the Dip in the years 1883-4 and 1884-5 was and reenwich , and and at Kew .
The differences in the Declination change are still more marked , as the following Table shows .
It will be noted that the annual variations differed in 1880 by , and the average difference irrespective of sign is It does not therefore appear advisable to reduce the observations to one epoch by means of the annual variations as determined at a single observatory .
On the whole a better result will probably be attained if we collect all the evidence at our disposal and deduce from it an average secular change .
The comparison of our own observations with those of Mr. affords the means of this quantity for Scotland , and proves that the rate of decrease of the Dip is less in the northern than in the southern stations , a result which is in accord with previous observations .
The mean decrease in Scotland between 1837 and 1857 was .
( ' Brit. Assoc. Rep 1859 , p. 169 .
) In the nearly corresponding interval between 1837 and 1860 it was on the northern border of England and on the south coast .
( ' Brit. Assoc. Rep 1861 , p. 260 .
) tely very few Dip observations were made in England and none were made in Ireland in the 1857 survey , but are a number of stations common to the surveys of 1837 and 1886 .
We can thus determine the secular decrease during the last 50 years .
In the following Table the stations are grouped according their geographical distribution , the districts being North and South Scotland , North , Central , and South England and Wales , and North and South Ireland .
The data for the 1837 survey are taken from the paper by the late Sir EDWARD ABINE , already cited ( ' Phil. Trans 1870 , Vol. 160 , p. 271 ) .
SURVEY OF BRITISH ISLES FOR EPOCH JANUARY 1 , 1886 .
In the case of twelve Scotch and six English stations we are able to compare the rate of change in the two periods 1837-1857 and 1807-1886 .
The data are taken from the sources above described .
The mean ratio for North Scotland is , for South Scotland , and for England These figures prove that the ratio of the mean annual decrease , in the intervals 1837-57 and 1857-87 has been nearly constant all over Great Britain , but that the rate of change has been rather more rapid in the north .
If we assume that the ratio has been for England and Ireland , we may deduce , the mean annual decrease between 1857-87 , from the corresponding quantity given in Table II .
, p. 82 , for the interval 1837-87 , by the formula The same is determined for statiollS in Scotland by -irect experiment by the comparison of our own observations with those of Mr. WELSH .
The results are given in the following Table .
MEAN Annual Decrease of Inclination in Scotland between 1857 and 1884-8 .
Using these figures and the numbers calculated as above described from p. 82 , we get the mean annual decrease between 1857 1886 in different parts of the United Kingdom , as follows:\mdash ; we give these numbers as showing the mean results for districts of siderable size , the figures obtained at the individual stations appear worthy of careful study .
The observed values of the annual decrease for 1857-87 for Scotland , and those calculated for the same period for land and Ireland from the 1837-87 surveys are inserted in the map ( fig. 1 ) .
The general increase from east to west is evident , but in Great Britain there are some stations the annual decrease at which is less than at those which are nearest to them on the eastern side .
To test whether the differences were accidental a number of supplementary stations were taken on the Wye .
At Malvern and Brecon the calculated secular rates between 1857 and 1886 are and , and at Clifcon and-Cardiff\mdash ; 87 and .
In each case SURVEY OF THE BRITISH ISLES FOR THE EPOCH JANUARy 1 , 1886 .
'85 the smaller value occurs at the western station .
Between them lie Hereford , Ross , and Chepstow , and the rates obtained from observations at these in 1889 , are , and , which are in all cases less than those at Malvern and Clifton .
These observations then support the view that there is a real followed by a minimum rate of change of Dip in this neighbourhood .
Secular change of Dip , 1858-86 .
This induces us to show in fig. 1 that if all stations at which thele is a maximum rate of change are joined , we get two long lines running from Edinburgh to Exeter , and from Stornoway to Limerick .
The stations are not sufficiently numerous to enable us to draw any certain conclusions about them , but the point ought not to be lost sight of when the survey is repeated .
For the present we note only that of three pairs of neighbouring abnormg stations at which observations were made in all three surveys , all three are abnormal ( in the sense that the rate of change is greater at the easterly station ) whether we take the interval 1837-87 or 1857-87 .
In both cases the rate at Inverness is greater than at Fort Augustus , at Glasgow than at Cumbrae , and at Berwick than at- dinburgh .
The distribution of these stations affords , at all events , prima facie evidence that the differences between them are not accidental .
That they cannot be due to errors of measurement is evident when we remember that an error of in the annual rate of change mined from periments separated by an interval of fifty years would require an error of 5 ' in one or other of the observations , or of in both .
We therefore decided to regard the figures in fig. 1 as giving the true rate of secular change of Dip between 1857 and 1887 at the stations indicated .
The determination of the mean rate of decrease of Dip between the years 1857 and 1887 does not , however , give the value of that quantity during the period of our survey .
As the rate is diminishing it be less in recent than in earlier years .
The evidence we are able to collect on this head is confined to data from Greenwich , Kew , and Stonyhurst , and to a few stations at which we observed both near the beginning and near the end of our survey .
In the following fable we have entered such facts as were known to us at the time when the progress of the reductions made it necessary to come to a decision on this point .
Our own observations and those taken at Stonyhurst are corrected for diurnal variation and disturbance .
The illustrates the importance of these corrections wJlen conclusions are to be drawn from observations between which the interval of time is small .
These figures illustrate the fact that there may be a considerable discrepancy at any particular station between the actual secular change at a given time and that deduced from the mean taken over a long series of years .
On the whole , we think it probable that the present rate of change of Dip is somewhat less than its mean value between 1857-87 , and we have , therefore , reduced the numbers for that period by one tenth , neglecting fractions of a tenth of a minute .
This assumed gives a correct mean rate for the stations referred to above in Table III .
, though it is considerably too high for some and too low for others .
It is probable that in the case of Kerrera the observed rate of change is not quite correct .
The site is subject to local SURVEY OF THE BRITISH ISLES FOR EPOCH JANUARY 1 , 1886 .
attraction , and a single pair of observations made at so short an interval as a year is hardly sufficient to rely on .
The results from Stonyhurst , however , prove that the assumed and actual rates of change may differ by , and that , thus , Dips reduced to January 1 , 1886 , from the end of 1888 , may be about wrong .
To prevent , as far as possible , any effect on the general run of the isoclinal lines from this uncertainty , we put down on a map all the assumed rates places for which they could be determined , and took the rate at any other place to be that given by the nearest of these stations .
The rates of decrease thus obtained for the south of England are considerably less than those assumed by M. MOUREAUX for the north of France , which do not differ much from .
These values are obtained for the interval 1858-1885 by comM .
x 's with LAMONT'S map .
We think , however , the observations of the Rev. S. J. reduced to the epoch , January 1 , 1869 , prove that in France as in England the rate of change of Dip has been diminishing .
Twelve stations were common to ather PERRY and M. MOUREAUX ; at one of these , Marseilles , the station of the earlier observer was no doubt seriously affected by local disturbance .
The rate of change for the epoch is much larger than for any other station , and for 1869-85 it is much oo small .
Rejecting this observation we think the following table shows that M. MOUREAUX'S nul1lbers are boo large .
We turn next to the secular change of the Declination .
A number of stations are common to our own survey and those of 1837 and 1857 , and we have also observed at several places where the Declination was determined in 1872 .
The mean annual decrease calculated for the observations is exhibited in the following table:\mdash ; TABLE IV.\mdash ; Table of Mean Annual Decrease of Declination .
SURVEY OF THE BRITISH ISLES FOR THE EPOCH JANUARY 1 , 1886 .
These results indicate\mdash ; ( l ) that the secular change is greater in the northern than in the southern parts of the United Kingdom ; ( 2 ) that it was much larger between the years 1857-86 than during the interval 1837-57 ; and , lastly , ( 3 ) that , in the north at all events , the rate of annual decrease is again diminishing .
The first of these conclusions differs from that arrived at by Sir E. SABINE when coilating the 1837 and 1857 surveys .
He assumed ( ' Phi ] .
Trans vol. 160 , 1870 , p. 268 ) " " an annual decrease of West Declination of approximately in Scotland and the north of England , increasing to in the middle and souihern parts of ngland and to 6 ' in Ireland Our conclusion that , at the present time , the rate is greater in the north is borne out by the values deduced by M. MOUREAUX from his own observations and the maps of LAMONT ( loc. cit. , p. 162 ) .
The values he gives f'or stations on the anean and on the English Channel are about and respectively , which indicates an increase of about on the rate for each degree of latitude as we go north .
The truth of the second conclusion is rendered more ) arent by calculating the annual rate for the epoch instead of 1857-86 , as shown in the following Table for the Scotch stations , for which the requisite data are available:\mdash ; MEAN Annual Decrease of Declination .
The third conclusion is supported by a comparison of M. MOUREAUX'S rates for the epoch 1859-1885 with those obtained by collating as before his those of the Rev. S. J. PERRY ( loc. cit This result\mdash ; that the rate of decrease is diminishing\mdash ; is in accord with the conclusions drawn by M. MOUREAUX himself from a comparison of his own observations with those on which a magnetic map of France for the epoch 1875 was based by M. MARIE'-DAVY ( loc. cit. , p. 166 ) .
He did not , however , make use of this fact , as the number of observations employed by MARIE-DAVY was small , and they were taken at stations irregularly distributed over France .
He does not cite the observations of Father PERRY , which , as the above Table shows , strongly support the same view .
In the next Table we give such observations as were made at English observatories or by ourselves during the progress of the survey which bear on the question under discussion .
No very definite conclusions can be drawn from them .
Two single observations at the interval of a year , as at Kerrera and Glasgow , are hardly sufficient for the purpose of dedu ing a secular rate .
Loch Aylort and loway are both disturbed stations and the results are thus less trustworthy .
The general evidence that the secular change is greater in higher latitudes is opposed by the fact that Stonyhurst gives a less than Kew and Greenwich .
If the increase with latitude which obtains in France were maintained in England , the value at Paris being taken at , we could deduce for London ; for Edinburgh ; and Wick ; of which the two latter are less than the values for the epoch 1872-86 given in Table , p. 88 .
SURVEY OF THE BRITISH ISLES FOR THE hPOCH JANUARY 1 , 1886 .
From Table IV .
we see that the rates at Greenwich for the three intervals , 1837-86 , 1857-86 , and 1872-86 are nearly the same , the differences between them and the mean rate during the period of our survey being , and respectively .
Assuming that if we subtract these numbers from the rates calculated for same epochs for the other English stations iu Table IV .
we obtain the present rates at the stations , we find as a mean rate , which agrees closely with that of Greenwich .
On the whole we decided to take a secular rate in England south of a line joining Redcar and Barrow , and also in Wales , and in Ireland south of a linejoining Dublin and Donegal .
For the remainder of England and Treland we have assumed the rate to be per annum .
At the time when most of the Scotch observations were reduced we had not M. MOUREAUX'S results before us , and we hardly felt justified in departing from the rates given by the comparison of our observations with Mr. WELSH'S , on the evidence of the small number of stations at which the Declination was determined in 1872 .
We , therefore , employed rates varying from in the north of Scotland to in the south , the change on the east coast being somewhat greater than on bhe west .
These may be a little too large but the error certainly does not exceed the variation which occurs between neighbouring stations .
For the later Scotch observations we used a rate The data for the determination of the secular change of the Horizontal Force are more scanty than in the case of the other elements .
At the time of the 1837 survey the measures were comparative only , and though Sir E. SABINE afterwards reduced them to absolute values by means of the known absolute value and secular rate at Kew , we do not think any useful end would be attained by discussing them .
M. MOUREAUX found that for the interval 1848-85 , the secular variation in France attains its maximum ( M.U. ) at Bordeaux and decreases slowly towards the $ From information kindly supplied by the Astronomer-Royal .
east , being0025 at Paris and at Nice and hfezieres ( loc. cit. , p. 167 ) .
The mean value obtained by the Rev. S. J. PERRy in 1869 by comparison with LAMONT was ( ' Phil. Trans Vol. 160 , 1870 , p. 48 ) , which was increasing by about .
annum and would correspond to at the ent time .
By arison of our own observations with those of WELSH we obtain as a mean for Scotland between 1857 and 1886 .
The following Tables give these values and those which were obtained at the English observatories or by ourselves during our survey .
Secular Change in Horizontal Force , .
SURVEy OF BRITISH FOR THE CPOCH JANUARY 1 , 1886 .
9.3 The results at Kew and Greenwich show that the secular variation varies so mnch from year to year that it is practically impossible to draw any conclusions from so short a period as 2 or 3 years , and our own observations on Scotland show that if to this difficulty , due to a slight variation in the position in a disturbed district is added , the results are still less trustworthy .
At the time when the reductions were made the Greenwich esults for 1886 and the Stonyhurst results for 1887 had not been published .
We therefore took the mean of the values for Stonyhurst 1883-1886 , Kew 1883-87 , and Greenwich 1883-1885 , or the annual increase for England .
This value is in fair accord with M. M.ouREAUX'S results .
In Scotland we took the number given by the comparison of our own and WELSH'S observatioIjS , viz. , , and for land Results of the Observations .
Having thus described the observations and the methods of correction and reduction to epoch , we now proceed to give a more detailed account of the results at each station .
In doing this we have attempted to distinguish between facts which it is necessary to give for theinformation- of most of those who may read our paper , and details which ought to be preserved , but which will nevertheless only be of interest to observers who may for any cause wish to undertake a detailed examination of our results .
We have therefore decided to publish in this paper only a description of each station , the hours at which the observations were taken , the results corrected for diurnal variation and disturbance , and the mean reduced to epoch , As Ma , netic Observatories and provincial Colleges are now rapidly multiplying , it is not too much to hope that studies of the districts in their immediate neighbourhood may from time to time be made by those connected with such institutions .
It is , therefore , we think , important that the descriptions of the sta , tions should be readily accessible .
In adding to these only the times at which the observations were taken , and the actual and reduced results , we are publishing far less than has been usual .
Thus , in the case of the Rev. S. J. PERRY'S survey of the east of France , he gave for the Force observations the date , hour , and temperature for both vibrations and deflections , the time of one vibration , , the distances of the magnets in the deflection experiment , the observed deflections , and .
M. MOUREAUX has not given quite so many details of the Force observations , but in the case of the Declination he gives the individual readings for the determination of the geographical meridian , \amp ; c. , so that the description of his work at each station occupies about two quarto pages .
For our own part we have no fault to find with the publication of these details ; on the contrary , we have found them to be useful ; but the large number of stations included in our survey would , we fear , make this paper inordinately long if we adopted a similar plan .
We therefore purpose to place in the hands of the Royal Society bound copies of the details of the observations and calculations , and also of the forms described on pp. 71-74 , in which the results are analysed and the corrections for diurnal and secular variation and disturbance are applied .
It will thus be possible for those who may desire to do so to inspect these , and the data used in the preparation of the Tables on pp. 77-80 will be on lecord , while this paper will , we hope , contain sufficient to enable future observers who are not specially interested in the details of the calculations and reductions to find the positions where we observed , to know when we observed there , and tojudge from the final results of the general accuracy of the observations .
We have followed the plan adopted by M. MOUREAUX of giving all the facts with respect to each station together , which we think the most convenient .
Tables are also given on pp. 251-258 in which the final results are entered in tabular form for comparison with the values obtained by calcu- lation from formulae to be hereafter discussed .
The stations are arranged in the following order:\mdash ; Three groups are formed , comprising Scotlan England and Wales , and Ireland , this being the chronological order of the bulk of the observations in each of these countries .
In each group the stations are arranged in alphabetical order , and they are numbered continuously throughout .
These numbers are affixed to the positions of the stations as given on Plate T. , which serves as an Index map .
The Scotch stations , from Aberdeen to Wick , are numbered from 1 to 54 , and it should be mentioned that the name of a Loch is regarded as deterrnining the initial letter .
Thus East Loch Tarbert is found under T. The glish and Welsh stations , including the Isle of Man and the Channel Isles , from Aberystwith to Worthing , are numbered from 55 to 156 .
The Irish stations , from Armagh to Wicklow , are numbered from 157 to 200 .
Thus anyone desirous of looking up the observations at a particular place , can easily do so from a knowledge of its name , while the stations in any particular district can be found in Plate I. , and then referred to by means of the corresponding numbers .
The data given in each case , are as follows:\mdash ; ( 1 .
) The number and name of the station .
( 2 .
) Date of observations .
Initials of the observer and numbers of the instruments .
( 4 .
) Latitude and longitude of the station .
( 5 .
) Verbal description of the station .
For the Declination we give : ( 1 .
) The time from the souihing of the sun ( 2 ) , at which the geographical meridian was determined by sun observations , a positive sign indicating the afternoon .
( 2 .
) The G.M.T. of the determination of the magnetic meridian .
( 3 .
) The observed Declinations with all corrections applied ( 8 ) .
( 4 .
) The mean observed Declination reduced to the epoch , January 1 , 1886 .
MR. A. W. RiJCKER AND DR. T. E. THORPF , ON A MAGNETIC 99 .
KEW .
Lat. ; Long. In the Magnetic House at the Observatory .
Declination .
It will be observed that the differences between these results are not exactly in accord with those which would be deduced from the table on p. 58 .
This is due to the fact that in the above table the observations have been treated as though Kew were an ordinary station , whereas , on p. 58 , the corrections have been applied in the most accurate way .
In the first place , no attempt is made on p. 58 to reduce the observations to epoch , SURVEY OF THE BRITISH ISLES H'OR THE EPOCH JANUARY 1 , 1886 .
161 whereas in the above table they are reduced by the mean coefficient of secular change at Kew during the period of the survey , together with a mean correction for monthly variation .
In the next place in the reduction on p. 58 , no distinction was made between the diurnal variation and disturbance , the total divergence from the mean being read off directly .
The corrections applied in the above table were obtained by reading off the disturbances only and aPterwards adding the value of the diurnal variation .
The differences are a little larger than we should have expected , but this is probably due to the fact that , as the corrections were wanted quickly , the constants of the curves were only determined provisionally .
That the accuracy attained was practically sufficient is proved by a comparison of the numbers in the above table with those on p. 58 .
The mean difference between our results and those given by curves treated as accurately as possible is , while the mean difference between the numbers given above and their mean is The Kew value for January , 1886 , is , and ours , from the above table when reduced to the Kew instrument by adding the mean difference , is If from each of our uncorrected observations given on p. 58 , we subtract the difference between the corresponding Kew value and , we reduce to January , 1886 , by one operation , in which no distinction is made between disturbance and the diurnal , monthly , and secular variations .
largest of the numbers so obtained is , and the smallest , while the largest and smallest in the above table are and .
Thus , not only are the mean results obtained by the two methods in close accord , but the range of variation of the results is in each case practically identical ; in other words , they agree to within the limits of the error of experiment .
Horizontal Force .
Inclination .
In the following Table the values for January 1 , 1886 , deduced above from the survey instruments , are given in Column I. In Column II .
are the mean differences between the survey and Kew instruments ( see p. 58 ) , and hence in Column .
the Kew ( calculated ) values are deduced .
In Column IV .
are the published elements for that epoch ( ' Roy .
Soc. Proc vol. 41 , 1887 , p. 416 ) , the Declination being corrected for diurnal variation .
The agreement between .
and Iy .
is satisfactory in the cases of the Declination and Dip .
In that of the Horizontal Force the actual value in January , 1886 , seems to have been rather high .
As a good deal will hereafter be said about the relative values of the element , at Kew and Greenwich , and as we have not compared our instruments with those at Greenwich , it has been thought better to treat both these observatories in the same way , and therelore to use the numbers in Column IV .
as the Kew values .
SURVEY OF THE BRITISH ISLES FOR THE EPOCH JANUARY 1 , 1886 .
TION 0F THE ISOMAGNETIC LINhS .
( 1 .
) The Lines .
The isogonals , isoclinals , isodynamics and lines of equal Horizontal Fotce are drawn through points , at which the values of the magnetic elements to they refer are equal .
A general term which shall include them all would be useful , and we venture to suggest as the most obvious and convenient .
In nothing do different magnetic surveys differ more widely than in the methods eraployed of these lines .
Some observers have calculated them by least squares ; others give maps on which they are exhibited , but say nothing about the principles i11 accordance wit which they have been drawn .
But , inasmuch as the main object and result of a survey is the delineation of these isomagnetics , it seems to us that it is most important that they should be drawn with all the accuracy that the observations will allow .
Before proceeding to discuss the method of doing this , it will be well to consider the exact to be attached to the operation , and to attempt to give greater precision to the language in which the various curves , and the physical phenomena which they represent , are described .
If we suppose that a series of magnetic curves are drawn , in which all distortions due to local magnetism are neglected , except which are on a scale comparable with the dimensions of the earth itself , they would be the gnetic lines ; on the other hand , lines which showed every disturbance , howeyer large or small , would be the gnetic l The object of a survey is to determine as nearly as possible the forms of the true lines , and to deduce from them the directions of the terrestrial lines in the district under investigation .
Between these two extremes various grades of accuracy of detail intervene , and the terrestrial lines may be regarded as affected with distortions of different orders due to disturbances of various .
It is of course impossible to frame definitions which shall accurately distinguish between them , but it nevertheless convenient to recognise three classes , into they may be divided with respect to any particular We may regard a disturbance as .
of the third , second , or first class according as its range is less than the ayerage distance between the stations , greater than this distance , but small compared with the dimensions of the entire area under survey , or such as to involve the whole or a considerable fraction of that area .
The term local may be reserved for disturbances of the third ) , which affect only a single station or its immediate neighbourhood , and are represented by minor bends or small loops in the isomagnetic curves .
Those of the second class may be called regional disturbances .
They are represented by considerable distortions of the curves , but do not seriously interfere with the determination of the general direction or average distance apart of the terrestrial lines .
A disturbance of the first class is however an obstacle to any inference as to the relation between the true and the terrestrial curves .
It may tort them similarly , and so lead to false conclusions as to their undisturbed directions , or it may introduce such widespread and important irregularities that any conclusion deduced from the data afforded by the survey would be manifestly uncertain .
It is to be noticed that the sa1ne cause may produce disturbances of different classes in the lines corresponding to different elements .
Thus a broad band of magnetic rock at right angles to the magnetic meridian , the extremities of which lay outside the district under investigation , would produce little effect on the Declination , but might affect the Dip to a much larger extent .
If an equation can be found to a family of curves which represents , as closely as possible , the general direction of the true isomagnetics of a particular kind throughout the whole country , the curves are the nearest approach to the terrestrial lines which can be deduced from the observations .
The undulations of the district lines on each side of these are evidence of regional disturbances , while the still more sinuous lines obtained by joining the points calculated as corresponding to any particular value of the element from the values in two neighbouring stations are influenced by the local disturbances also .
If the district under survey is very small , the assumption that the terrestrial lines are is very approximately true , but the nature of the curve indicated by a straight line on a map depends on the projection on which the map is drawn .
The English observers have generally determined the position of the station by its distance north or south of a particular line of latitude , and east or west of a particular meridian , both distances being expressed in geographical miles .
That this system is open to objection is evident by considering its application to a simple ideal case in which the geographical meridians are themselves supposed to be the isogonal lines .
It would evidently be better under such ditions to take the latitude and longitude as coordinates , and inasmuch as in the neighbourhood of the United Kingdom there is an approximation to such an arrangement , it is probable that the latitude and longitude are at least as convenient as any other data for the determination of the position of che stations .
By this plan also , we are saved the trouble of converting the longitude or west of Greenwich into miles east or west of the prime meridian .
We have therefore taken the latitude and longitude as our ates .
In the next place it is , we think , important to find the general equations to the isomagnetic lines , but it would be difficult to determine their form with sufficient accuracy by a graphical process .
We therefore decided to trace them by a preliminary process of calculation which was carried out as follows .
The country was divided into nine overlapping districts bounded by lines of latitude and longitude .
Ifthe stations within any district were not uniformly distributed , they were weighted , so that the weighted number of stations per unit of area should be everywhel.e about the same .
In speaking of the mean value of any quant tie a SURVEY OF THE BRITISH ISLES FOR THE EPOCH JANUARY 1 , 1886 .
235 district , it will be understood that the values appropriate to any station are throughout properly weighted .
The means of the latitudes : gitudes , and declinations determined the central station of the district and the declination at that station .
It was then assumed , if and be the district of a station , i.e. , the differences between its latitude and longitude and those of the central station , that the declination is connected with these quantities by the linear equa , tion\mdash ; Two equations of condition were then formed by adding the equations thus obtained\mdash ; ( l ) , for all stations to the north of the central station ; ( 2 ) , for all stations to the east of it , and dividing by the ) of stations employed , each multiplied by its proper weight .
By solving these for and , the rates of change of the Declination per degree of latitude and longitude respectively were obtain To test this method , it was applied in the case of the Dips to the whole of Scotland .
This portion of Great Britain furnished a severe test , as from the irregularity of its form it is not particularly well adapted for the application of the method of equations of condition .
The calculation was also made by the method of least squares .
Both calculations were repeated twice , viz. , with and without the inclusion of Soa and Canna , at which the local disturbance is very consirdeable .
The central station was not that given by the mean latitude and longitude , but that obtained from WELSH'S Dip observations , viz. , N. ; long . .
In this particular , the conditions of the selected example were unfavourable to the method of equations of condition .
In the following Table , is the ] made by the lines of equal Inclination with the geographical meridian , is the change in Inclination ( expressed in minutes ) per geographical mile , measured at right angles to the ] inals , and is the Dip at the central station:\mdash ; plan very similar to that above described was employed by Dr. VAN RIJCKEVORSEL in working up the results of his survey of the Tndian Archi e ( ' Magnetische Opnemiug van den Indischen Archipel in de Jaren 1874-77 , gedaan door Dr. ' Amsterdam , J. MULLER , 1879 ) .
MR. A. W. RUCKER AND DR. T. E. THORPE ON A MAGNETIC These results prove that in this case , at all events , the differences bebween the results of the two methods of calculation are not greater than those produced in the numbers given by the method of least squares according as stations affected with considel.able disturbances ( and 76 ' respectively ) , and to 4 per cent. only of the total number , are included or excluded .
We do not think , therefore , that it is in general advisable to use so cumbrous a method as that of least squares , when the addition of a station 0 two may modify the results to an extent far exceeding the error with which numbers obtained by the equations of condition are likely to be affected .
If , however , the district under investigation is of such a shape that the effects of change of latitude and longitude respectively cannot be easily separated , it may be desirable either to modify the rule for obtaining the equations , or to employ least squares .
In the case of a district so large as Sc.otland there is another objection to the use of least squares , , that the fundamental assumption on which that method is based is almost certainly not true when applied to it .
We shall show hereafter , as is indeed already known for other districts , that the errors are not irregularly distributed over the entire area , but that large fractions of the whole are affected with errors of a particular kind .
We cannot , therefore , regard the employment of least squares as theoretically loetter , while it is certainly practically more inconvenient than the method of equations of condition .
The following Table contains the boundaries of the nine districts , the latitudes and the longitudes of the central stations , the values of the Declination at the central stations , and of the change in Declination per degree oflatitude and longitude ( and dS'/ d ) both expressed in minutes of arc .
TABLE V. In District I. , on account of its irregular form , the method of equations of condition is not very table , and the method of least squares has been used .
In order to compare this with the formula obtained by STEWART from WELSH'S observation the SURVEY OF THE BRITISH ISLES THE EPOCH JANUARY 1 , 1886 .
237 same coordinates have been employed as he used ( the -mile system ) , hence the values of and can only be considered as applying to the central stations .
By means of the first the Declination at any point on the meridian through that station can be calculated , and for other points on the parallel of latitude passing through such a point , the formula must be employed where ?
/0 and refer to the central station .
all the other districts the values of and are valid for the whole district .
By means of these formulae the Declination was calculated for all points within the United Kingdom defined by whole degrees of longitude and half degrees oflatitude , e.g. , for , long .
E. , 1o E. , W. , and so on .
Where the districts overlapped the means of the numbers thus obtained were taken .
All these values are given in the following table .
The figures in brackets at the end of a row indicate the number of the district from which it was deduced .
Where two or more districts overlap , the individual ations are given in italics and the mean in rdinary type .
Throughout the central parts of the kingdom the agreement between the numbers given by the linear formula proper to different districts is sufficiently close to leave little doubt that the meam cannot be.more than 1 ' or 2 ' wrong .
Where greater differences appear it is generally easy to account for them .
Thus , lat. long .
W. is in the highly disturbed region in the west of Galway .
long .
, is close to a remarkable and hitherto unsuspected disturbance in the eastern counties , of which we shall have more to say hereafter .
The large differences on the border of Districts I. and lat. are , perhaps , in part due to the irregularities in the shapes of these districts owing to which the formulae are not obtained under favourable conditions .
Both the other elements , however , agree with the Declination in indicating violent local disturbance in this region , and there can be no doubt that the discrepancies are due to a physical cause .
TABLE VI.\mdash ; Declinations Calculated MR. A. W. RUCKER AND DR. T. E. ON A MAGNETIC Isogonal lines were next drawn by the aid of the mean values of the Declination given in this able , the points at which they intersect any particular line of latitude or longitude being calculated on the assumption that the rate of change of the Declinatiun with latitude or longitude may be regarded as constant over a single degree .
The curves thus obtained are shown in dotted lines in Plate II .
As they were deduced from the linear district formulae they may be called the district curves .
By drawing a smoo curve to coincide as nearly as possible with any one of the longer of these broken district curves , a close approximation to the corresponding isogonal could be obtained , but in order that the intervals between the curves might be properly spaced out , it was thought better to obtain a general formula by which they could be expressed .
It appears from ] V. , p. 236 , that increases with the latitude , but is nearly independent of the longitude .
values of are more irregular .
After sevel'al trials ib was found that the mean Declinations at the central stations of the various districts could be reproduced very accurately by means of the formula , , where and are the numerical values of the latitude and longitude expressed in degrees and fractions of a degree .
It must be distinctly understood that this formula has no theoretical value except in so far as it expresses satisfactorily the equation to smooth curves drawn according to a definite rule to represent the general form of the broken district curves .
In applying it to the various stations the curves given by the periodic terms were drawn , and the values corresponding to the latitude and longitude of each read off .
It is possible that errors of or may have occurred in this process .
The mean values of the Declinations at the central stations of each district are in the following Table compared with the results given by the formula .
SURVEY OF THE BRITISH ISLES FOR THE EPOCH JANUARY 1 , 1886 .
241 In District III .
we included Stonyhurst ; in District IV .
Greenwich , Kew , and Berck.sur-Mer , and in District VII .
, Cherbourg .
data for the French stations were those given by M. MOUREAUX , reduced , of course , to January 1 , 1886 .
The values of the constants in the equation were not finally chosen until the Declination had been determined by means of a preliminary formula osely areeing with that given above for all the points indicated in Table VI .
, pp. 238-9 , and the values given by it and by the district curves compared .
Over the greater part of the country the agreement is extremely close .
It is not necessary to reproduce the numbers here , as in Plate II .
we have ] down both the broken district curves , and also the smooth curves given by the equation .
The former are dotted , the latter are continuous lines .
The agreement is all that can be desired , except , perhaps , on the coast of Norfolk , and on the west and north coasts of Scotland .
Taking , however , the English Channel first into consideration , we have calculated the points at which our , etc. , isogonals cut lat. .
When secular change is thus allowed for , these correspond to the positions of the isogonals on January 1 , 1885 , i.e. , with those of M. MOUREAUX .
We have also measured the corresponding points from his map .
The result is shown in the following Table:\mdash ; The isogonal does not pass through France , and , therefore , M. MOUREAUX'S map is not of special authority on its direction , but there seems no doubt that while the isogonals cut latitude at the same point , there is a considerable divergence between the others .
According to the general formula , is the change of Declination per degree of longitude on lat. The two Channel Districts in Table V. , p. 23 and IV .
agree in giving for this number , and if we take this as correct and adhere to 55 ' W. for the isogona ] , we obtain the following values:\mdash ; NIR .
A. W. RUCKER AND DR. T. Hi .
THORPF ON A MAGNETIO These are in much better accord than those obtained from the general formula , and the conclusions which may be drawn from them are that in latitude ( 1 .
) M. MOUREAUX'S lines are certainly not too far to the west ; ( 2 .
) And our isogonal is not too far to the west .
Both these conclusions are important .
M. MOUREAUX believes that there is what we should call a considerable regional disturbance in Brittany .
The isogonals drawn by him sweep out to the west in the western part of France and do not resume their normal course until they have reached the English coast .
In the English Channel , therefore , they are deflected to the west , and as both our district lines , and to a much more marked extent our terrestrial lines are still further west , it is evident that M. MOUBEAUX has not in any way exaggerated the westerly tendency of the lines in the western parts of the Channel , and that , therefore , our observations tend to confirm his view .
It is probable that our terrestrial curves are in this district a little too far to the west , but it must be remembered that lat. is almost outside the region of our survey , and that in the Channel Isles , which with Cherbourg , are our only stations to the south of it , there are considerable disturbances .
In fig. 2 the continuous lines are our terrestrial isogonals , which on January 1 , 1886 , corresponded to M. MOUREAUX'S , etc. , isogonals on January 1 , 1885 , the secular change being about 7 ' .
M. MOUREAUX'S curves are shown with dashes and dots , and the hypothetical connections in dots .
A satisfactory comparison on the ders of the areas of two surveys cannot , however , be made unless there is a closer agreement in the methods of working up the results of the observations than there is between our own and M. MOUREAUX'S .
Coming next to the we find that on the coast of Norfolk the calculated Declinations are generaIIy much lower than the observed values .
This would be remedied by moving the further to the east , but the fact thab it is already to the east of M. MOUREAUX'S line is a strong argument against such a course .
It is noticeable that Sir FRED .
EVANS appears to have found some difficulty in this part of the country Phil. Trans ) .
His curves are shown in Plate II .
The points at which the isogonals cut lat. are given below , and it will be seen that the distance between them is a maximum in the centre of England .
MR. A. W. RUCKER AND DR. T. E. THORP ON A MAGNETIC he assumes in the western parts of the English Channel is continued 10 the Straits of Dover , his isogonal and own would be in practical accord .
As regards the considerable deviation between the district and calculated terrestrial lines in the North of Scotland , we have no hesitation in adhering to the calculated lines as being probably the more correct .
The masses of basalt and other igneous rocks which occur in the West of Scotland and North of Ireland produce a very marked effect on the magnetic elements .
The general formula gives the Declination at the outlying station at Lerwick almost a , s accurately as the local district equations obtained from Scotland only .
The three values Observed Declination . . . . .
Calculated by Regarding this station as giving a iixed point it is evident that in drawing the curves to it from the South of Scotland , great weight ought to be given to the general law which is found to -be obeyed with accuracy in Great Britain from the English Channel to the Tay , and in the South and East of Ireland .
The only way of eliminating the effect of regional disturbances of such magnitude as these which exist in Scotland , is by studying the shape of the terrestrial lines in adjacent districts .
Observations to the north of Scotland would be very valuable for this purpose , but as it would be difficult to obtain them we think that the results of the general fonnula must be accepted as a close approximation to the truth .
We shall return to this subject in the discussion of local disturbances .
( 2 .
) The Isoclinal Lines .
The isoclinals were obtained in a precisely similar manner to that above described .
The districts were the same , but the positions of the central stations were in some cases slightly different , as at a few places Dips had been observed without Declinations and vice , versed .
In District VII .
the inclusion of the Channel Isles led to rates of change with latitude and longitude so widely different from those obtained elsewhere that it was thought better to omit them .
When this was done the coefficients assumed normal values .
We give in the next Table the latitude and longitude of the central stations , and the corresponding values of the Dips , and the rates of change of Inclination per degree oflatitude and longitude .
SURVEY OF THE BRITISH ISLES FOR THE EPOCH JANUARY 1 , 1886 .
245 TABLE VII .
A general formula was next found to embrace the whole .
For this purpose a Table similar to Table VI .
, pp. 238-9 , was prepared , and the district isoclinal lines were drawn from it on curve paper , on a purely artificial system , in which all degrees of latitude and longitude were regarded as of equal length .
When thus drawn the mean directions of the isoclinaIs were nearly straight lines and practically parallel .
The equation to the isoclinal was where and are the latitude and longitude expressed in degrees and fractions of a degree .
If then be the length of the perpendicular on this line from any point , the Dip at the station indicated by that point would ( if the lines were equidistant ) be given by the equation where A is a constant .
This condition was not fulfilled .
The distance between the lines increased approximately in the proportion of their distance from the line , but a small periodic term was necessary in addition to this correction to produce the desired accuracy .
The following plan was finally adopted:\mdash ; If we write and we get the Dip in degrees from the equation The Dips given by this formula for the central stations in each district are , in the followin able c ared with those given in Table VII .
, p. 245 .
In PIate III .
we show the broken curves obtained by the district lines and also the terrestrial isoclinals as represented by the formula .
The agreement is on the whole satisfactory , but it is possible that a closer approximation to the terrestrial lines might have been obtained had we made the inclination to the geographical meridian increase rather more rapidly in the west .
rfhis would have diminished the discrepancy in the south of Ireland .
On the other hand it would have considerably increased it in the north of Ireland and the central districts of Scotland , in which large regional disturbances undoubtedly exist .
It does not , however , appear to be safe to depart from the rule that the terrestrial lines deduced from any survey should give , as nearly as possible , the mean directions of the true lines in the district under investigation .
This end is better attained by our formulae than if the agreement were closer in the south of Ireland .
Our isoclinal agrees almost exactly that of M. MOUREAUX in the more easterly parts of the English Channel ( see fig. p. 243 ) .
Both just cut Dungeness and Beachy Head .
For the longitude of Falmouth , however , our line is 10 or 12 miles to the south of that of M. MOUREAUX .
As it is at this point much nearer to the English than the French coast , our result is probably the more trustworthy , and this opinion is confirmed by a study of the map of the French Survey .
The isoclinals drawn by M. MOUREAUX upon a map on Mercator 's projection are curved in the south of France , the convexity being towards the north .
The curvature becomes less as the Dip increases , and the line is represented as quite straight .
M. MOUREAUX cannot have much to guide him in drawing this line , and a divergence such as that which exists between his line and ours could easily be introduced by an apparently trifling error in the estimation of the rate of disappearance of the curvature of the lines .
Our lines would be slightly concave to the north if drawn upon a map on Mercator 's projection .
SURVEy OF THE BRJTISH ISLES FOR THE EPOCH JANUARY 1 , 1886 .
247 ( 3 .
) Lines of Horizontal Force .
These lines were treated in the same way as the isogonals and isocIinals , and in the following Table the constants for the nine districts are tabulated as before .
As Professor BALFOUR STEWART did not deduce the lines of equal Horizontal Force from Mr. WELSH'S observations , there was not as much reason in this case as in the others for taking the whole of Scotland as District I. We therefore included in this district only all stations north of lat. By means of these data a Table like Table VI .
, pp. 238-9 , was prepared , and lines of equal Force were drawn on curve paper as in the case of the isoclinals ( see p. 245 ) .
These at once showed that the lines above and below that corresponding to units were differently disposed .
The mean direction of each of the southern lines be accurately represented by a linear function of the latitude and longitude , their departure from parallelism is not great , and their mean distances are nearly the same .
On the other hand , though the lines inDistrict I. , i.e. , in the extreme north of Scotland , are parallel to the line , which runs from the neighbourhood of Miltown Malbay to that of Scarborough , the intermediate isodynamics make smaller angles with the geographical meridian , and their average distance is greater than in the south .
It is , however , difficult to decide what is the direction of the terrestrial lines .
Too much weight must not be attached to the fact that the lines in the north of Scotland agree with those in the Midlands and south of England , as they are deduced from a single district of irregular form and the seat of great local and regional disturbances .
There is also reason to suppose from what is known of the lines of equal Horizontal Force on the continent that the distance between them increases towards the east , and those crossing England are so represented on M. MOUREAUX'S map .
It may well be , therefore , that the diverging lines are those which agree most closely with the terrestrial lines of equal Horizontal Force .
On the whole then , the risk of introducing fictitious disturbances by attempting to include the whole country under one simple law would be very great .
We decided , therefore , in calculating the lines of equal Horizontal Force , to employ different formulae for districts to the north and south of the line .
Taking the southern district first , the lines may be regarded as but , inasmuch as they run across the whole breadth of the kingdom , a very small error in the calculated slope is important .
A rather complicated formula is therefore unfortunately necessary .
If be the mean distance , expressed in degrees of latitude and measured along longitude W. , between any line and that which corresponds to units , we have a relation of the form where is a constant .
The line cuts longitude W. in latitude , and thus the equation to the isodynamic through is where measures the slope of the line .
Neither nor are quite constant , but both are functions of , so that ; .
Thus , if we wish to find where the line corresponding to cuts longitude , we find from the equation where the value of , used in the small term in , is the approximate value obtained by neglecting it .
From this is found , and is then known .
If the latitude and longitude are given , and the Horizontal Force is required , we first find approximately from the formula .
Subtracting from this , the difference is , whence is found .
Taking next the district north of the line , there is no particular difficulty in finding similar equations to express the mean direction of the lines with great exactitude .
If we write SURVEY OF THE BRITISH ISLE EPOCH , all that can be desired in this respect is attained .
These equations do not , however , give the forces at the central stations with the accuracy attained in the southern districts .
Far better results are obtained if we assume that the lines are parallel to the line and are at equal distances from each other .
In this case is constant , and , in addition to the first of the above equations , we have only The values obtained from the central stations by both formulae are given in the following Table : \mdash ; We have , therefore , to choose between two formulae .
The simpler represents the lines hout Scotland as practically parallel , both with those which the south extend right across the kingdom without considerable curvature and with those found by the district equations for the of Scotland , and also reproduces the values of the Horizontal Force at the central stations with great accuracy .
The more complex expression represents the mean directions of the lines in the centre of Scotland and north of Ireland more satisfactorily , though these are obviously affected by some widespread disturbance , but it fails when tried by the test of the reproduction of the forces at the central stations .
We think there can be no question that , under these circumstances , the simpler formula , which assigns directions to the in harmony with those obtained in other parts of the kingdom , must be that selected ; and we have therefore employed it .
In the following Table the values of the Horizontal Forces at the central stations are compared with the calculated values given by the formulae\mdash ; MR. A. W. RIiCKER AND DR. .
THORPE ON A MAGNETIC and for values of ] by and for values of The difference between the observed and calculated results in District VIII .
( South Ireland ) , is larger than in the other cases .
It must , however , be remembered that it is absolutely very small .
Thus , if the disturbed stations at Galway ( which have been excluded from the calculations were introduced , the difference would be reduced from 0022 to 0007 .
As the effect of one station is so great , it is remarkable that the discrepancies are not larger .
We conclude this section with a Table , which gives the observed and calculated elements at every station .
In the calculations , graphic methods have been partly used , and , in some cases , this may have to slight differences ( which are , however , so small as to quite unimportant ) between the values given and those which would be deduced directly from the formulae .
SURVEY OF THE BRITISH FOR THE EPOCH JANUARY 1 , 1886 .
259 LOCAL AND REOIONAL URBANCES .
We now come to the consideration of the local and regional disturbances which exist in many parts of the United and which we have investigated much more fully than has hitherto been done .
As the problem is one of difficulty , it is necessary that every step should be carefully considered .
It may be attacked in three ways , which are not independent , but each of which is attended with special advantages and disadvantages , and which , when combined , afford in many cases the means of arriving at definite conclusions .
( 1 .
) lf the true isomagnetic curves are drawn as accurately as possible , without any attempt to smooth the irregularities , they present in disturbed districts distorted forms which enable us to judge of the nature and magnitude of the disturbance .
The great advantage of this method is , that it is independent of calculation .
It is not affected by errors possibly introduced by the method of determining the terrestrial curves .
The conclusions arrived at are based directly upon the observations .
The objections that may be raised to it are , that it can only be used with effect in the case of considerable disturbances , and that it may tend to exaggerate the importance of those which , though of great local intensity , are of small range .
A curve may be carried many miles from its position , in order to pass through a station , the disturbance at which dies out within a very short distance .
When the stations are as numerous as ours , and when the curves are drawn with a due regard to the possibility of isolated maxima and minima , we do not think that the risk of error on this account is as great as it might at first sight appear , but in so far as it exists , it may be checked by the second method .
( 2 .
) If the disturbances of the elements , i.e. , the rences between the observed values and those calculated from the general equations given above , are plotted down on a map , it is found that they are not scattered haphazard , but that certain districts exhibit definite peculiarities , such as that the observed value is always too large or too small .
From a study of such maps deductions can be made as to the nature of the disturbing forces .
This method is open to objection that the peculiarities in question may not correspond to physical realities , but may be due to the inadequacy of the formulae .
The calculated rate of increase of the declination wich longitude , for instance , may be made a little oo rapid in one part of the country , and a little too slow elsewhere .
The observed declinations may , herefore , appear too Jarge in the one district and too small in another , and thus a mere error may create a fictitious attractive or repulsive force .
To this objection , it may be answered that the observed disturbances are too large and too irregular to be thus explained .
Taking , for instance , the English and Welsh stations which.lie between long , 3 W. and long .
5 W. , the mean disturbances of the Declinations for the groups indicated , are as follows:\mdash ; Four most northerly stations W. Intermediate group of eight stations Six most southerly stations . . .
W. A glance at the isogonals ( Plate II .
) in this part of the kingdom , is sufficient to show that ic is impossible to believe that the real terrestrial curve not only crosses and recrosses that which have drawn between Plymouth and Holyhead ( which is quite ossible ) , bnt that the amplitude of the oscillation amounts to more than 20 ' of ation , or to about 45 ' of itude .
If this were so , the terrestrial would be nearly as uous as the true curves ( cf. Plate V It is also to be observed , that even if there is a slight tendency of the kind supposed , it will be partly corrected by the first method .
The objection to that is , that too mucb stress may be laid on the peculiarities of individual stations\mdash ; to this , that too great weight may be given to the characteristics of districts .
If the two methods point to the same conclusion the two criticisms are mutually destructive .
A more serious objection is , that the indications are , at times , somewhat ambiguous .
An.increase in the Declination may be due to a small force acting at right angles to the magnetic meridian , or to a large one acting nearly parallel to it .
The value of the disturbance of the Horizontal Force will often decide to which of these the effect is due , but nevertheless , it is not easy to deduce definite conclusions from a mere inspection of the maps .
This consideration has , therefore , led us to supplement this method by another .
( 3 .
) In this third method , we calculate the magnitude and direction of the disturbing force at each station .
If S be the observed , and the calculated value of the Declination , if a similar notation be used for the other elements , and if , and be the northerly , westerly , and vertical components of the disturbing force , we have S , whence the magnitude and direction of the disturbing force are known .
Obviously the danger to be feared , in this case , is that the differences with which we deal are too small to give trustworthy results .
The calculation must , however , lead to conclusions compatible with those to which the study of the disturbances points , and even if tho results are only approximate , they are valuable as indicating , in a way which is more easily interpreted , the direction and magnitude of the forces under investigation .
Having thus described the general course which we propose to adopt , we shall now consider the methods more in detail .
After that we shall apply them to certain districts , ofwhich we have made a special study , and then to the whole area of the survey .
In attempting to discover the order which underlies the apparent irregularity of the disturbances , it is necessary to proceed on a working hypothesis , and we shall SURVEY OF THE BRITISH OR THE EPOCH JANUARY 1 , 1886 .
261 postulate only the possibility of the existence of points or surfaces which exert magnetic forces .
They will be called attractive or repulsive aocordin as they attract or repel a north-seeking pole .
If the existence or apparent existence of such centres is established the cause of the phenomenon will be a proper subject for enquiry .
The distortion produced by a symmetrical mass of netic matter on the isogonals depends upon the angle which these curves make with the netic meridian .
If we are dealing with a sufficiently small area , both the isogonals and the magnetic meridians may be represented by straight lines .
If these lines are mutually perpendicular , and if the declination is westerly and increa , ses the latitude , the forms of the true isogonals in the neighbourhood of a attracting mass , the centre of which is below the surface at , will be of the type sIIown in Fig. 3 , in which NS is the magnetic meridian through A. onals ruuuing approximately cast and west .
\mdash ; True produced by centre of attraction at A. In Japan the distribution of the isogonals and meridians approximates to this simple arrangement , and in the neighbourhood of the Fossa Magna according to Dr. NAUMANN ( ' Die Erscheinungen des Erdmagnetismus , ' Stuttgart , 1887 ) , a very remarkable disturbance of this kind is produced .
If the onals are nearly coincident , and if the Declination increases with the longitude , the effect on the isogonals of a weak attractive centre will be of the type shown in Fig. 4 .
If , however , the disturbing force is sufficiently powerful to make the Declination , at some point to the east of the centre , greater than its undisturbed value at the centre , the form of the lines must approximate to that shown in Fig. 5 .
The Declination will increase rapidly with the longitude , attain a maximum value , fal ] to its normal value at points on the magnetic meridian which pass through the centre\mdash ; on each side of which the attracting matter is supposed to be symmetrically situated\mdash ; then fall to a minimum , and finally resume its normal rate of increase .
If , retaining all the other assumptions , we suppose that the meridians are inclined at an angle of less than , the distorted curves will assume forms imilar to those in Fig. 6 .
It will be noticed if the undisturbed isogonals be SURVEY OF THE BRITISH ISLES FOR THE POCH JANUARY 1 , 188 263 characteristics may be reproduced , and a knowledge of their forms is very useful in interpreting maps on which the true isomagnetic curves are drawn .
These facts may also be illustrated by means of diagrams of another kind .
Thus if , in figs. 7 and 8 , the ] of the line AD represents the normal rate of increase of the Declination with longitude , the effect of an attractive centre below A will be represented by the curves shown , which correspond to the two cases of a weak and strong attraction respectively .
If , however , the force is repulsive , the curves assume the forms shown in figs. 9 and 10 .
In this case there may be two maxima and minima , following each other in order on opposite sides of the centre of repulsion .
The isogonals will be drawn together in its neighbourhood , and there may be two systems of loops , one on each side of the centre .
A centre of attraction will also be indicated by a convergence of the lines of equal Horizontal Force and Dip in its neighbourhood .
The observed will be greater than the calculated Force to the south , and smaller to north , of the centre .
The Dip , on the other hand , will be less and greater than its calculated values at stations to the south and north respectively of a point at which the directions of the normal and disturbin Forces coincide .
The phenomena are more complicated if the effect of the disturbance is to increase the value of the element on side of the centre of attraction on which it would be normally the greater , instead of increasing it ( as in the case of the Declinations ) on tha , t side ( the east ) on which it is normally the less .
Thus , if we represent a northward movement by progress along the line WE ( fig. 10 ) from to , the effect of a strong centre of attraction would be to check the ordinary decrease of the Horizontal Force with latitude .
If the centre were sufficiently powerful , the decrease might be converted into an increase ; and sinoe immediately over the centre value must be normal , a minimum and maximum must follow in order .
Another minimum and maximum may occur on the further side .
The effect of a centre of attraction on the lines of equal Horizontal Force is thus the same as that of a centre of repulsion on the isogonals .
In the case of the Dip , as in that of the Declination , there can only be two critical points on opposite sides of the point defined above .
The Vertical Force will increase with latitude at more than the normal rate to the south of a centre of attraction , and at less than the normal rate to the north of it .
The lines of equal Vertical Force will thus be drawn southwards in the neighbourhood of a centre .
If the Vertical Force attains a maximum value it will be at a point above the centre , and it must be followed by a minimum\mdash ; at which , however , the Vertical Force will be greater than its calculated value .
It is evident from this discussion that there are a number of signs of a centre of attraction which may not all coexist , and which will be complicated in actual practice by irregularities in the distribution of the attracting masses , but which may nevertheless be of considerable practical help in a survey of local magnetic disturbances .
( Cf .
LAMONT : 'Erdmagnetismus in Nord-Deutschland , ' 1859 , p. 21 .
) We have dwelt on them , not because they present any difficulty , or even because they altogether novel .
Mr. BENNETT BROUGH , A.R.S.M. , has given a full account of the methods employed by the Swedish mining engineers in exploring for iron ore .
Use of the Magnetic Needle \amp ; c. 'Journal of the Iron and Steel Institute , ' No. 1 , 1887 , pp. 289-303 .
) They are accustomed to map out the neighbourhood of a mass of ironstone with a magnet , in order to determine its exact position .
We believe , however , that the systematic use of the forms of the isomagnetic curves has been largely overlooked in the case of surveys comparable with that described in this paper .
In Plates V. to VIII .
, the values of the Declinations , Horizontal Forces , Dips , and Vertical Forces determined at the various stations are entered and true magnetic curves are drawn .
These curves have not been chosen for equal differences of the values of the element to which they refer , but those have been selected which exhibit the most marked peculiarities .
In this way attention is best drawn to disturbed stations and districts which can then be studied in accordance with the plan above suggested .
Turning next to the investigation of the disturbances or differences between the observed and calculated values of the elements , we note that apart from the assurance they afford that the peculiarities studied are common to a district and do not depend only on a single station , they often ) additional information to that which can be gained from the true isomagnetics .
Thus , in the case of the Horizontal Force , a maximum followed by a minimum may be either to the north or the south of the centre of attraction .
In the former case , SURVEY OF THE BRITISH ISLES FOR THE EPOCH JANUARY 1 , I88 265 however , both the maximum and minimum values are less , while in the latter case they are greater than the normal values at the stations at which they occur , and if one maximum and minimum only are formed , the disturbances at once decide whether they are to the north or tlth of a centre of attraction .
The lines which separate regions of positive disturbance in which the observed is greater than the calculated value from those in which it is negative are also of importance .
Thus , in the case of the Horizontal Force , if such a line is approximately perpendicular to the magnetic meridian , and if in passing over it from south to north we leave a region in which the disturbance is positive and enter one in which it is negative , we either pass over a centre or line of attraction or pass from the range of the influence of one centre of repulsion into that of another .
If the centre is not at a great depth below the surface , and if the magnetic matter is not widespread its minimum depth , it will cause a sudden reversal in the of the disturbances which will , however , be large near the centre on both sides .
On the other hand , from the region of influence of one centre to that of another ill , if the distance between them is considerable , be marked by a transition from small positive to small negative disturbances , or vice Similar remarks may be made with respect to the disturbances of the Declination , but they are most easy to interpret when the line which separates a positive from a negative runs approximately north and south .
If such a line meets another which separates a southern region of positive from a northern region of negative Horizontal Force disturbance , all the disturbing forces in the hbourhood t towards the point of intersection .
It will be convenient to speak of such a point as a peak , and to call a line which divides regions of positive from of negative disturbance of the Declination or Horizontal Force , so as to indicate attraction towards it , a ridge line .
In like manner , a line which separates the regions of influence of two attractive centres may be called a valley line .
In the choice of these terms we are , no doubt , influenced by our views of the facts to be hereafter set forth , but they are convenient , quite apart from any theory of the cause of local magnetic forces .
It will be proved beyond doubt that in some cases these forces emanate from matter below the surface of the earth .
If this is so , an increment in their intensity must be due either to a closer approximation to the matter or to an increase in its magnetisation .
We have no magnetic test to discrimate between the two , and therefore , without prejudice , adopt a nomenclature which is perhaps most consistent with the first hypothesis .
If we wished to keep absolutely free from all expression of an opinion as to the causes of the phenomena , we might have called a peak ( to which the lines of magnetic disturbing force converge ) , a magnetic sink , and so on , and for the present the terms we suggest may be taken as merely points and lines from and to which such lines of force run .
The disturbances of the Declination and Horizontal , together with the ridge and valley lines are shown in Plates IX .
and X. MR. A. W. RUCKER AND DR. T. E. ON A MAGNETIC The absolute value of the disturbance of the Vertical Force is especial]y uncertain .
In the cases of the Declination and Horizontal Force a centre of attraction affects stations on opposite sides with disturbances of opposite signs , to that the true mean value for the district can be found , and the true disturbance at each station deduced .
It is , however , possible that the disturbance of the Vertical Force may always be of the same .
Thus , if the centres of force are rocks magnetised by induction , the north-seeking pole of a magnet would most frequently be attracted downwards in the northern hemisphere as the north-seeking poles of the rock magnets would be deeply buried in the earth .
This hypothesis is that which experience has justified in Sweden .
If this is so , the mean value of the Dip or Vertical Force in a district be greater than its undisturbed value , and negative values do not necessarily indicate an upward force .
It is therefore safer only to use the disturbances of the Vertical Force as a means of indicating relative maxima and minima .
If , however , centres of repulsion exist , they might be detected by the observation that a large negative ( upward ) disturbance of the Vertical Force was accompanied with a sudden reversal of the disturbing Force in neighbouring stations .
If the Horizontal Forces are small , the negative disturbing Vertical Force is more probably to be interpreted as indicating a downward attraction of less than average nitude .
On turning to Plate XI .
, in which the disturbances of the Vertical Force are plotted down , it will 1oe observed that there is , on the whole , an excess of positive values in the south-east , and that large negative values are more common in the north and west .
It is obvious that in Scotland the disturbances are much greater than in , and very large positive values of vertical turbance occur in the western isles .
Indeed , if we include the enormous value obtained at Canna , the average for the whole of Scotland is positive .
Leaving out , however , this very abnormal station , it can hardly be doubted that negative values are more common in the north-west .
If this result could be trusted as corresponding to physical fact , it might indicate that the country as a whole is magnetised in the direction of the magnetic meridian .
We cannot , however , draw such a conclusion , more especially as it can be shown that the observed effect may probably be a result of our ignorance of the datum lines from which the disturbances of the Dips and Vertical Forces to be measured .
We have taken the Dip , as given by direct observation , as one of our fundamental elements , and in drawing the terrestrial curves which satisfy the conditions that the mean value of the Dip for each district when attributed to the central station is accurately reproduced by the formula , we have been compelled to ignore the possibility of unbalanced downward Vertical Force acting at every .
If we suppose that in consequence of this the calculated Dips ought , all over the country , to be diminished by a positive quantity , we get for the decrement in Vertical Force SURVEY OF THE BRITISH ISLES FOR THE EPOCH JANUARY 1 , 1886 .
267 If we take the values of and at Wick and St. Leonards , and write for the circular measure of 1 ' multiplied by , the unknown number of minutes by which the calculated Dip has been taken too large , we get at Wick .
St. Leonards Hence in the north the error will be greater than in , the south , that is , the excess of the calculated over the true undisturbed Vertical Force is greatest in the nortl ] .
Consequently the difference between the observed and calculated Vertical Force will be more frequently negative in the north , the excess of the error amounting to metric unit for every minute by which the calculated Dip is too large .
If , then , rt from local attractions the , as a whole , attracts the needle so that the Dip is everywhere 10 ' greater than it would be if the British Isles were replaced by sea , the error in the calculated Vertical Forces deduced from calculated Dips , obtained on the assumption that no such defect exists , would be greater in the north of Scotland than in the Channel , so that negative disturbances would largely predominate in the north .
As from the geological character of Scotland it is probable that the error in the calculated Dip would not be constant , but would be greater than in England , the validity of this explanation is even more probable this ation indicates .
It is not , therefore , safe to draw conclusions based on the relation between the Vertical disturbing Forces at distant stations .
In the calculation of the disturbing forces we not aiming at , and could not attain to , results which would give more than a general idea of their direction and magnitude .
It may be possible to conclude with certainty that an attractive mass exists near certain stations , even if the .
directions of the forces are wrong by or and their magnitudes are inaccurate by 50 per cent. To show that a degree of accuracy of this kind is attained we collect here the results at stations where we have taken full set of observations at two or more places , or on different occasions .
There are of course localities where it would be hopeless to get a good result from a single observation .
Such places are Canna and Portree .
At these we should never have observed except for the sake of investigating the disturbances which we knew were very great , and we therefore do not now take them into account .
It is sufficient if the agreement between different obseryations is satisfactory at stations which we regarded at the time of observation as normal .
At Storxloway we made four sets of measurements , the first in 1884 in the Castle Grounds , the other three in 1885 and 1888 on Ard Point , about a ] distant from the former station .
The ground was not good , as when we obseryed simultaneously about 50 yards apart on Ard Point the declinations differed by 18 ' .
At Loch Aylort we observed in 1884 and 1888 as nearly as possible on the same station .
At Oban we observed on the mainland in 1888 , and on the island ofKerrera MR. A. W. RUCKER AND DR. T. ] ] .
THORPE ON A MAGNETIC which is about a mile distant , in 1884 and 1885 .
In the Sound of IsJay we had two about four miles apart , viz. , Port Askaig in 1884 and Bunnahabhain in 1888 .
All these places , though typical Scotch stations , were on ground which elsewhere would be considered but indifferently good for our purpose .
' At three other places nlOl.efavoul.ably situated we also observed twice , , at Stranraer in 1884 and 1888 , at in 1886 and 1888 , and at Bude Haven in 1886 .
At the two former the stations were as nearly as possible the same on the two occasions , at Bude Haven they wers les3 than ]lalf a mile apart .
The esults are summed up in the following Table .
The resultant Horizontal disturbing Force is indicated by , and the angle which.its direction makes with the geographical meridicm by .
The latter is taken as positive on the western side , due north being epresented by The annexed small map ( fig. 11 ) illustrates these numbers by showing the direction and magnitudes of the disturbing forces as determined on two occasions at stations in Scotland .
An inspection of this map and of ) table justifies the statement that the magnitude and direction of the disturbing forces can be determined with an accuracy sufficient to us to draw conclusions from groups of stations even if it woudd not always be safe to argue one .
SURVEY OF THE BRITISH ISLES FOR THE EPOCH JANUARY 1 , 1886 .
Fig. 11 .
If the station is on good ground we may place more confidence in the results .
Stranraer is interesting as being the only place where we both took complete sets of observations simultaneously on good ground .
The results are in very close accord .
But those obtained elsewhere leave no doubt as to the order of the magnitude of the disturbing force nor as to its direction to within ( in unfayourable cases ) or The difference in the signs of the Vertical Forces obtained at Loch Aylort and Stornoway in different years may in part be due to the uncertainty of the secular correction for the ) , appears to be very abnormal , especially at Loch Aylort ( see p. 86 ) .
fn this case we should be driven to the conclusion that a real change in the local Force had taken place .
It is noticeable that the Vertical Force disturbance was apparently ( alg.ebraically ) greater at Stornoway , Loch Aylort , Kerrera , and Stranraer on our second visits to these places .
The following Table contains the particulars as to the disturbing force at every station in accordance with the notation described above .
On Plate XIIT .
, the directions and magnitudes of the Horizontal disturbing Forces are shown , and regions of positive and negative Vertical Force disturbance are cated , the former being shaded . .
A. W. AND DR. T. E. ON A lfAGNETIC 0F SELECTED DISTRICTS .
Having described and , as we hope , justified the methods by which we propose to investigate local and regional disturbances , we now proceed to discuss their application to districts of which we have made a special study .
The results we shall arrive will help us in the further ] ucidation of the magnetic state of the whole country .
The Hills .
The general nature and direction of the magnetization of igneous rocks is a problem on which comparatively observations have been made .
It is , indeed , known that when examined in detail they present great irregularities , and Commander CREAK , F.R.S. , has shown Proc that when islands disturb the magnetic needle in the northern hemisphere they attract , and in the southern repel the north-seeking pole of a .
This is what would be expected if they were the upper extremities of magnetic masses gnetized by the Earth 's induction .
like manner in Sweden , where ( as has already been stated ) the method of searching for iron ore by means of the magnet has been carried to considerable perfection , the assumption made , and justified by experience , is that the ) parts of the beds of ironstone attract the north-seeking pole .
Observations somewhat similar to those of Commander CREAK can be carried out on land in cases where igneous rocks rise in the midst of sedimentary deposits , and such observations are specially interesting in cases , such as that of the Malvern Hills , in which the axis of the magnetic mass runs north and south .
If its depth is considerable with respect to its length , and if it is magnetized by induction , we should expect the upper visible parts to attract ( in the northern hemispbere ) the north-seeking pole .
If , on the other hand , it is possible to conceive of a shallow mass of magnetic rock surrounded by non-magnetic matter , the northern end might repel the north-seeking pole .
Finally , if the mass was itself magnetized independently of the present inductive action of the Earth , as is certainly the case with small masses of highly magnetized its effect on a compass needle could only be determined by experiment .
With the view , then , of making a beginning towards the study of these questions in the United kingdom , we determined the polarity of the northern end of the Malvern Hills .
This range consists of syenite and granite .
The mass of igneous rocks runs due north and south for about eight miles , and is the broadest part but little more than half a mile wide .
On its eastern side is a great fault which extends many miles north and south of the range itself Two stations , Great Malvern and Malvern Wells , were taken on the eastern side of the hills .
They were both on the Red Marl , and distant about mile and a quarter from the centre of the range .
Two SURVEY OF THE BRITISH JSLES FOR EPOCH JANUARY 1 , 1886 .
27 corresponding stations were also taken on the western side at about the same distance , and both on the Old Red Sandstone .
Of these Mathon was mile to the north of the latitude of Great Malvern , and the line joining them cuts the range at about a mile from its nort , hern end , passing close to its ighest point .
Colwall is about three-quarters of a mile to the south of the latitude of Malvern Wells , and the line joining them cuts the range about miles south of the northern termination of the igneous rocks .
The Declination and Force were detel.mined at all these stations .
Time did not allow of the Dip taken elsewhere than at Great Malvern .
The results are given in the following Table From these the disturbing forces were deduced , the notation being the same as that used on p. 268 .
The accompanying Map shows the direction of these forces and their relative magnitudes .
The Worcestershire Beacon 1440 is the highest poin on the range ; to the south of this the height diminishes , and then increases again to the Herefordshire Beacon .
The directions of the disturbin forces tend towards these hills , and the results are , we think , only compatible with the view that the Malverns the north-seeking pole of the magnet .
The attraction exerted by the Malverns having been demonstrated , it is convenient to discuss in the next place a locality where the disturbances are enormously greater .
Popular tradition has long attributed to the basaltic rocks of the Island of Canna the power of deviating the needle through very large angles .
The compasses of passing ships are supposed to be affected by the eastern extremity of the island , on which stands Compass Hill .
Magnetic observations were made on the Island by Sir FREDERICK EVANS ( ' Phil. Trans 1872 , vol. 162 , p. 325 ) , but , we have , we believe , been able to add considerably to what was already known of its magnetic properties .
The island , which is about fifteen miles south-west of Sky , is about five miles long , its greatest lying nearly due east and west .
It is divided into two approxi- mately equal portions by a neck of comparatively low elevation .
The highest ground is in the eastern part where it rises to a height of 724 feet , and our observations have been confined to portion .
The cliffs on the north side are here some hundreds of feet in height and sheer into the sea .
On the south side several small valleys lead to the shore , but the hills rise very steeply about a quarter of a mile inland , and the tops of several consist of irregular masses of columns .
On Compass Hill , SURVEY OF BRITISH ISLES FOR POCH JANUARY ] , 188 277 which is the most easterly point on the , the longer axis of the mass runs north and south .
Its height on the -nland side is from ] 20 to 40 feet , while it falls steeply towards the sea .
On another hill which rises behind Kaill , the residence of R. THOM , Esq. , the columnar mass is and more clearly defined .
The north or landward side is 15 or 20 feet high , the south side is a steep cliff We roughly estimate the length at about 50 yards .
In fig. 13 the position of this hill is indicated by , that of the , highest point on bhe island by , and that of Compass Hill by C. Tn 1884 we determined the magnetic elements by means of the Kew Magnetometer No. 60 , and the Dip Circle No. at the position marked P. The differences between the observed and calculated values are given in the following Although these observations show that the station was highly disturbed , the effect on the Declination is not so great as to be detected , except by a careful observation .
In 1888 we observed by means of a small azimuth compass at 23 stations which are indicated on the map , and determined at each the bearings of a number of distant points .
The observations were made on August 16 and 17 , bright sunny days with a northerly wind , on which the atmosphere was very transparent .
We were therefore able to take the bearings , not only of prominent headlands on Sky and Rum , but also of some distant objects , such as the Ushinish Lighthouse on N. Uist in the Hebrides .
The azimuth compass was a small instrument which did not admit of great accuracy , but , as will be seen in the sequel , it was sufficient for the purpose we had in We carried a chart with us , and marked the position of each station while on the spot .
On the average , about four were taken at each station , and at no station was the number less than two .
The bearings of the objects selected were afterwards taken from the chart , and , by comparing these with the observations , the Declination was determined .
The agreement between the individual observations at each station was in general only moderately good , but , even in cases where only two observations were taken , we think the means are accurate to about half a degree .
Throwing out thirteen stations which were evidently highly disturbed , we took the mean of the Declinations at the other ten as giving the mean Declination for the easterly half of the island , which is By subtracting from the Declinations found at the stations , we obtained the disturbance of the Dec]ination at each .
The results are given in the following Table :\mdash ; Stations XV .
to XVIII .
inclusive were taken round the basaltic mass on the summit of the hill behind Kaill , and on the sides indicated by the letters which follow the numbers .
Stations .
to XXIII .
were in like manner taken round the summit of Compass Hill .
In both these cases the compass was generally within a foot or two of the basaltic columns .
The observations , therefore , show that these are powerfully magnetic .
The disturbing force at Station XXI .
was nearly half that due to the horizontal intensity of the earth 's magnetic fieid .
On the other hand , it is evident that their influence diminishes very rapidly with the distance .
Station .
was only a few hundred yatds from the hill behind Kaill , yet the was not affected by more than Stations I. , II .
, III .
, and .
are grouped round the southern half of Compass Hill , at distances between 200 and 500 yards from the summit ; but the largest disturbance of the Declination is W. Stations XXIII .
and II .
are situated one above the other on the eastern side of Compass Hill , the horizontal distance between the two being not more than 80 yards ; SURVEY OF THE BRITISH ISLES FOR THE EPOCH JANUARY 1 , 1886 .
279 yet the disturbance diminishes from E. at Station XXIII .
, near the top of the hill , to E. at Station II .
, near its base .
These conclusions are completely borne out lloy ooservations made on the ' Coventina in which we visited the island in 1884 and 1888 .
On leaving Canna for Loch Boisdale in 1884 we sailed as close as possible to the northern face of the island , and took frequently the compass bearings of points on Sky .
We were unable to detect the smallest deviation of the needle .
In 1888 observations were made under still more favourable circumstances .
We approached the island from the north , and , when about three miles distant , the yacht was directed towards a mark on Rum , by which its course could be kep without reference to the .
compass .
We were then magnetic S. E. , in the most favourable direction for the effect of Compass ( if any ) to be detected .
We passed it within 200 yards of the shore , but obseryed no deviation of the compass ; and we are quite certain th , if there was any , it was less than a point , i. e. , less than The net result of our observations is that the basaltic cliffs of Canna are powerfully magnetic , and may deviate the needle of a compass placed near them by about two points , i.e. , about , but that the effect diminishes very rapidly with the stance , and is inappreciable on a ship 's compass 200 yards from the base of the hill , to which tradition ascribes , and in which we have ourselves detected , the most powerful magnetic properties .
We have adopted as the mean value of the Declination at the less disturbed stations in Canna in August , 1888 ; this leads to for January 1 , 1886 , which is only 6 ' in excess of the calculated value .
This is interesting , inasmuch as Plate IX .
shows that neighbouring stations have Declination disturbances of opposite signs , and indicates that a line of no regional disturbance near to Canna .
Thus the four stations , Kyle Akin ( No. 35 ) , Soa ( No. 46 ) , Canna ( No. 14 ) , and Loch Boisdale No. lie very nearly in a straight line , and the disturbances of the Declinations vary continuously , being , and \mdash ; 4 9 , which proves that Canna lies near an attractive centre or ridge .
This is in harmony with the fact that the disturbance of the Vertical Force is positive , and is enormously amounting to , or about of the whole Vertical Force .
A comparison of the results obtained at lalvern and points very clearly to the otherwise probable conclusion that far-reaching effects are to be expected quite as much from the great mass and uniform magnetization of rocks as from their being highly nlagnetized .
The basalt of Canna is far more susceptible and more magnet than the Malvern syenite , but we doubt if a mile and a quarter from Compass Hill it would produce an effect on the Declination needle at all equal to that which we have shown is due at that distance to the Malverns .
The sterrl Counties .
A problem of considerable interest has to be considered in connection with the Eastern and South-eastern counties .
In the greater part of this district , the surface soil is such that it certainly can produce no marked effect upon the magnet , yet in this apparently " " good ground we have found magnetic disturbances of very wide range .
It is well known that the Declinations obtained at the Kew and Greenwich Observatories differ more widely than the difference of longitude will explain .
The difference is not so great as appears at first sight , as the Kew results as published are not corrected for diurI ) variation .
The published mean Dec:linations at these two stations in 1886 are , Greenwich , and Kew , of which the Kew result must be diminished by about 6 ' .
This gives a difference of 16 ' , which exceeds thab corresponding to the difference of longitude by 10 ' .
We are not aware that any attempt has hitherto been made to connect this discrepancy between the Declinations at these two important observatories with any regional disturbance in their neighbourhood .
The fact which led us to believe that they lie within the area of such a disturbance was that there is not only a small decrease in the Declination between Kew and Reading ( instead of an increase as the difference oflongitude requires ) , but that there is also a very small difference between Worthing and Ride .
The Declination at Reading should be about 20 ' greater than that at Kew , but the observation made there in 1886 proved that it is 1 ' less .
The Declination was again determined in 1888 the same spot near Reading , with the result that the Declination , when reduced to epoch , came out lower than before , thus increasing the discrepancy between the actual and calculated differences between the two stations .
In like manner , the diffel.ence of longitude between Ride and Worthing is equivalent to a change of 22 ' in the Declination , whereas the observed value at Ride was only higher than that at the more easterly station .
In order to investigate these differences more fully , we determined to run a chain of stations along the valley of the Thames , to observe at another series half-way between the Thames and the Channel , and to interpolate a station between Worthin and Ride .
The result is shown in the annexed diagram ( fig. 14 ) .
Horizontal lengths indicate the longitudes of stations and vertical lines the Declination .
Three groups of stations , such that all places included in each are of nearly the same latitude are taken , thus forming three lines crossing the whole of the south of England .
In the cases of mouth and Ride , which lie considerably to the south of the line which nearly passes through the other stations correctionoe of 14 ' and 7 ' respectively have been added .
Chichester is to the north of the line and its Declination has also been corrected by 7 ' .
It will be noticed that th three exhibit an anomaly on or immediately to the south These results make it certain that near the south of the Wash the Declination diminishes instead of increasing ( as at normal stations ) with the longitude .
The accompanying map shows the directions and magnitudes of the disturbing forces in this district .
They indicate a centre of attraction to the north of the line which joins Spalding and King 's Lynn .
Tloe eicestershire District .
Another series of local disturbances exists to the west of that which has just been described .
The facts which first attracted our attention to it were that at two pairs of stations , , Birmingham and Northampton , Leicester and Peterborough , the observed Declination at the more westerly was only about 2 ' greater than that at the more easterly station , though the calculated differences were as much as 35 ' in the first case , and 27 ' in the second case .
The observations indicated that in this district the isogonal lines run nearly east and west instead of nearly north and south , and we proceeded to investigate their more closely .
We thought that the anomaly was probably connected with the fact that in Charnwood Forest , which is not very distant from Leicester , igneous rocks appear upon the surface , and observations were made round this district , though we were always careful that our station should be on what was apparently good observing ground .
We have thus confirmed the existence of a great easterly trend in the isogonal lines , and though the magnetic state of the district appears to be complicated , and to require further investigation , we have also established several facts which will probably prove to be of fundamental importance in the solution of the problems connected with it .
Three of the most interesting stations are .alville ( No. 76 ) , Loughborough ( No. 109 ) , and Melton Mowbray ( No. 116 ) .
Loughborough and Coalville are both on the Red Marl , with alluvium near to the streams .
Between them lies Charnwood Forest , in which are masses of porphyry , greenstone , and syenite .
We should , perhaps , expect from the analogy of the Malverns that at these two stations the needle would be attracted towards the crystalline rocks .
This does not appear to be the case , or , as is more probable , the stations are too far distant to be affected .
At Loughborough , which is the more easterly station , the disturbance of the Declination is towards the , while at Coalville it is in the same direction , but only to the extent of About 12 or 13 miles further to the east is Melton Mowbray , situated on Lower Lias clay with argillaceous limestone at its base , yet a series of observations made here on April 22 , 1888 , gave a Declination disturbance of , i.e. , towards the west .
Unfortunately , the observation for the geographical meridian could only be made near noon , as the sun was invisible during the rest of the day ; but so remarkable did the result appear that the place was revisited on Apri130 .
Another SURVEY OF THE BRITISH ISLES FOR THE EPOCH JANUARY 1 , 1886 .
289 station on the other side of the town , about a mile and a half distant from the first , was chosen , and the new observation was made under favourable conditions at about 5 . .
The result indicated a Declination disturbance of ) , which was in close accord with that previously obtained .
At first sight , then , it appears that the peculiarities of the district might be explained by the hypothesis that a centre of force , powerful relatively to Charnwood Forest , exists somewhere between Melton Mowbray and Loughborough .
The disturbances of the Declination at these two stations are in opposite directions and of nearly equal amounts .
This view does not , however , correspond with the directions of the disturbing forces obtained at MeJton Mowbray .
If it were correct they should both have acted in nearly parallel directions towards the west .
As a matter of fact , at the first station the disturbing force acts nearly due west , while at the second it is only from south .
This indicates that the disturbance at Melton Mowbray is of a more local character , and cannot , in the manner suggested , be brought into relation with the oppositely directed disturbances at Coalville and Loughborough .
If , however , we turn from the Horizontal to the Vertical disturbing Forces we find that the of these various stations may be connected , and ] a magnetic map of the district can be drawn which may furnish the first rough outlines to which details may hereafter be added without rendering them substantially incorrect .
This explanation is based upon the fact that whereas in this part of England the disturbances of the Vertical Forces are for the most part positive , at four consecutive stations in this neighbourhood they are negative .
These stations are Coalville , Leicester , Kettering , and Northampton .
Whether the low Vertical Forces are due to the presence of a repulsive centre , or to a deep cleft or valley in the attracting mass , we cannot tell .
Indeed , the latter hypothesis might account for a repulsion if the mass of attracting matter which we have supposed to culminate in the Reading peak terminated abruptly on its northern edge .
For , if it were magnetised by induction it is possible that some of the lines of force might escape upwards to the surface , diminish the Vertical Force , and urge the north pole northwards .
Contour lines drawn as in the accompanying map ( fig. 21 ) , indicate the possibility of a very sudden rise in the magnetic matter from a line drawn Coalville , Leicester , and Kettering , to another which passes through Chesterfield , Nottingham , Melton Mowbray , Peterborough .
This view is supported by the fact that a ridge line , on passage across which the disturbance of the Declination runs from near Chesterfield to near Melton Mowbray .
If this view were correct , Melton Mowbray would have to be regarded as near the summit of an extremely steep peak , as the Vertical Force changed from to in the small distance between the two stations .
The direction of the Honzontal Forces would indicate a point a little to the west of both stations as the actual peak .
SURVEY OF BRITISH ISLES FOR THE EPOCH JANUARY 1 , 1886 .
291 The whole district may apparently be divided into four parts , of which the boundaries are indicated on the map by heavy lines .
The southernmost is the region of the Reading disturbance .
The tendency of the Horizontal Forces to act towards regions of high Vertical Force is unmistakeable .
At King 's Sutton the direction of the resultant appears to be affected by the region of Vertical Force to the north , near Kenilworth .
At Worthing and Ride the Horizontal Forces point direct to the Chichester peak .
At Purfleet and Southend the directions of the Forces are more southerly than the distribution of the Vertical Forces would have led us to suspect .
The most easterly district ; that of the Wash disturbance .
Near King 's Lynn the Vertical Force is great .
It is greatest at Tilney , which is the central station .
The Forces at neighbouring stations converge to a point to the north of this , and the Horizontal Force at Tilney itself is directed northward .
It is therefore likely that here , as in the case of Reading , the true peak , though , is not at the spot at which we found the greatest Vertical Force .
It probably lies to the north of it .
The central district is that of the Leicestershire disturbance .
Here the phenomena are more complicated , and we wish it to be distinctly understood that we think it probable that Melton Mowbray does not occupy the position of unique importance which our observations allot to it .
Nevertheless , we must point out that the hypothesis that a narrow ridge of attracting matter runs somewhat in the position we have assigned to it , is remarkably supported by the direction of the Horizontal Forces at Coalville , Loughborough , Leicester , and Wanton , which would all be explained on this hypothesis .
At Melton Mowbray the directions are , of course , indeterminate by means of the Vertical Force , and it is quite possible that the phenomena observed there may be due to some relatively small dyke , and not to an uprising of a part of a widespread mass of igneous rock .
Until this district is more fully suryeyed we are justified in adopting the view represented on the map , which is consistent with all the known facts .
There is , however , one station in the district , , Birmingham , which is not in harmony with the rest , as the direction of the Horizontal Force is toward the region of minimum Vertical Force .
Perhaps this indicates that this not connected , as we have supposed , with the larger region of low Vertical Force to the west , but that the two are severed by a district of high Vertical Force running from Birmingham northward .
Future investigation can alone decide this question .
The most northerly of the four districts is almost outside the region of our special surveys .
We only introduce it to show that there is a large region of high Vertical Force to the north of Melton Mowbray , and that , therefore , there is nothing anomalous in the northerly directions of the Horizontal Forces at Nottingham and Grantham .
If it be true that in the Leicestershire disturbance the attracting matter lies within limits , it is quite possible that at these stations the predominant influence may be that of the larger northern mass .
MR. A. W. RUCKER AND DR. T. .
THORPE ON A MAGNETIC On the whole , then , if we take the region bounded by the sea , by lat. and W. , which includes about 50 stations , we think the Horizontal ForceS unmistakeably to act toward the region of great Vertical Force .
There is one striking exception at Birmingham , and one or two more doubtful ones on the lower reaches of the Thames , and the rule must be construed subject to the obvious condition that a true maximum of Vertical Force , though probably near to , is not necessarily at the station at which we happen to have found the largest among the Vertical Forces we have measured .
Subject to these exceptions and to this proviso the rule holds good .
GENERAL RESULTS 0F THE INVESTIGATION 0F THE LOCAL AND REGIONAL Having described the results obtained in districts to which we have devoted special attention we now proceed to apply the same methods to the whole area of the survey .
In adopting this course we are fully aware that the number of our stations is not sufficient to enable us to speak with any certainty as to the details of the magnetic peculiarities of the districts we are about to discuss , and it is quite possible that we may have arrived at some conclusions which must hereafter be modified .
It appears to us that evcn under these conditions our work is much more likely to be useful if we what only professes to be a first rough sketch map of the ma , .
forces in play in the country than if we leave our successors to get what hints they can from observations which we ourselves have made no attempt to colate .
Even , therefore , if our conclusions were much less certain than we believe them to be we think it would be better to state them .
Fortunately , however , we are able to take up a much stronger position than this .
Our conclusions may be tested , ( 1 ) by the agreement of the results of the various methods of attacking the problem , ( 2 ) by the agreement of our results in Scofland with those which can be deduced from Mr. WELSH'S survey , ( 3 ) by the establishment of relations between the magnetic phenomena at stations scattered over wide areas , and ( 4 ) by the establishment of a between the magnetic and geological characteristics of various districts .
In all these particulars we venture to assert that they will bear investigation , and we cannot but believe that we have detected the main directions of the lines of disturbing magnetic force .
The method we adopt is as follows:\mdash ; We draw the ridge and valley lines ( see p. 265 ) which mark the centres and the boundaries of districts which are under the influence of a dominant locus of attraction .
We take each district bounded by two valley lines , and study it by means of the true isomagnetics of the disturbances and forces , and lastly , we discuss the relations between its .
and characteristics .
SURVEY OF THE BRITISH FOR EPOCH JANUARY 1 , 1886 .
293 The are probable consequences of the hypothesis that each district , bounded by two valley lines , is subject to an attraction tending towards the centre :\mdash ; ( 1 .
) Since the regional forces are weak near a valley line it is in such a position that we should expect the effects of such local forces as might exist to predominate , and discrepancies to occur more frequently than elsewhere .
( 2 .
) The centre of a district , in the neighbourhood of the ridge line , should be a region of relatively high Vertical Force .
It will be remembered that the ridge line is drawn by means of the disturbances of the Declination and Horizontal Force only , and , therefore , an agreement between position and that of a region of high Vertical Force is an independent confirmation of the accuracy of the theory that the forces in play in the district form a connected system , that is , they are regional and not merely local .
( 3 .
) The directions of the Horizontal Forces at points of maximum or minimum Vertical Force may be indeterminate in the sense that they cannot be deduced from the distribution of the Vertical Forces .
This point has already been insisted on .
( 4 .
) In crossing a valley line both sets of attractions in the region which it separates tend to produce a maximum and minimum of the Horizontal Force ( see fig. 10 , p. 263 ) .
If the southerly attraction were alone in play both of these would have values less than the normal .
If the northerly acted alone both values would be greater than the normal .
Hence , since the maximum will be to the north of the minimum it is probable that in producing it the effect of the northern abtrarjtion will predominate and the values will be greater than the normal , while in the case of the minimum they will probably be less .
Comparison with Mr. WELSH'S Survey of Scotland .
No more severe test of the physical reality of the sturt ) forces deduced by us from our observations can be applied than by the enquiry whether similar methods applied to Mr. WELSH'S survey lead to results .
In part , the severity of the test lies in the fact that the methods cannot be precisely similar .
As England was so imperfectly surveyed in 1857 , we are compelled to assume a function to express the isomagnetics in Scotland at that date .
The isogonals and isoclinals were calculated by BALFOUR STEWART , who used the geographical mile system of co-ordinates .
Unfortunately the third element selected by him for calculation was the Total Force , which throws but little light upon ] such as those we are now investigating .
We have , therefore , calculated the lines of equal Horizontal Force , using , as in our other calculations , the differences between the latitudes and longitudes and those of the central station as co-ordinates , and by combining these with BALFOUR STEWART'S calculated values we have found disturbing forces at the various stations .
Again , Mr. WELSH omitted the determination of one or more of the elements at many stations .
He apparently had chiefly the terrestrial isomagnet , ics in view , and did not consider that for the discovery of the local peculiarities of a station , it is essential that all three elements should be Jined at it .
In part this omission was due to misadventure .
A number of Declination observations had to be rejected as the mirror was found to have been out of adjustment .
If , therefore , the calculated magnitudes and directions of the disturbing forces depend to any great extent on the distribution of the stations or on the method of reduction , conclusions drawn from WELSH'S survey and our own could .
not be in harmony .
In the case of the Declination comparison is possible , not only with the survey of , but with the observations collected by Sir F. vANS in 1872 .
We have measured the distances of the stations from the isogonals given by him ( ' Phil. Trans 1872 , vol. 162 , p. 319 ) , and have deduced the disturbances .
In the accompanying map ( fig. 22 ) we have combined the results of all three surveys .
The direction of a short line drawn through the stations , shows whether the north pole of the needle was deflected to the east or west .
Observations made by Mr. WELSH are indicated by a dotted line , those due to the naval officers by two short lines with a dot between them , and our own by a continuous line .
A discrepancy between any two of the surveys is thus indicated by a cross .
Lines of no disturbance are drawrl , separating districts in which the regional disturbance is of opposite signs .
and passing , when possible , through stations at which different results have been obtained in different surveys .
Without insisting on the accuracy of the details of these curves , we think that a study ) the map can only lead to the conclusion that the districts which they bound are affected by some common cause which produces a similar effect upon the needle .
It can hardly be doubted for instance that the magnet is deviated in opposite directions in the neighbourhood of Elgin and Banff .
All three surveys agree as to this fact .
, the evidence is very strong that on the mainland to the north of the Caledonian Canal , except , perhaps , near Cape Wrath and the Pentland Firth , the needle is deflected by regional forces to the east , while in the Islands on the West Coast and the Mull of Cantyre it is deflected to the west .
Of course it would be possible to find in these districts places , such as Canna or the Cuchullin Hills in Sky , where large deviations in both directions could be obtained within a few yards , but the great majority of survey stations are chosen too carefully for the local error to become so overwhelmingly predominant .
We have omitted Portree because the ground there is known to be unfavourable to our purpose .
The differences between the three results obtained are too great to make a mean value trustworthy .
At Canna on the other hand , where it will be remembered we Qerved at twenty-three places , the mean value cleared of larger disturbances agrees with the results obtained at neighbouring stations .
The following facts are also important .
Seven stations were common to the.surveys of 1857 , 1872 , and 1886 .
One of DECLINATION Disturbances .
It is difficult to offer any satisfactory suggestion as , to the causes of the apparently egular change .
It may be due only to the fact that the three methods of deducing the disturbances by calculations are different , or it may be due to real changes in the local forres .
Such alterations are not impossible since if the rocks were magnetised by induction on the earth 's field the direction and intensity of the induced magetisation would be subject to secular change , but the differences are too great to be explained thus .
But whether local changes have taken place or not , there can be no doubt that the general distribution of the Declination disturbances in Scotland is the same now as it was in 1857 .
In Map 22 the dotted lines are the curves of no disturbance which would have been drawn from our own observations alone .
The differences introduced by the addition of WELSH'S and EVANS'S surveys are quite unimportant in the North .
In the South they simplify the lines instead of introducing complications .
As the disturbances at Edinburgh and Berwick are of opposite signs , we had to a line of no disturbance between them . .
WELSH , however , had two stations a little to the West of Berwick , , Makerstoun and Melrose , and at these the disturbance is of the same sign as at Edinburgh .
The introduction of these justifies us in treating Berwick as an isolated station , and the simple system represented by the continuous lines results .
We have adopted these as the true lines of no Declination disturbance in this part of the Kingdom .
As Sir FREDERICK EVANS did not add any Dips or Horizontal Forces to the list of those already known , we proceed at once to the direct comparison of the disturbin forces deduced from our own and Mr. WLLSH'S surveys .
The magnitudes and directions of these calculated by us as above described are as follows:\mdash ; MR. A. W. RtlCKER AND DR. T. E. THORPE ON A MAGNETIC distorted ( Plate V and indicate a centre of attraction between the two .
such points are marked Loops in the Vertical Force isomagnetics indicate maxima and minima of Vertical Force , which are marked and respectively .
As no observations could be made to the west of the islands in the south-west of Scotland , it is impossible to prove formally that the Vertical Forces in this district are maxima .
The values at Scarnish and Iona are , however , the same as those at Wick Golspie respectively , which are nearly two rees farther north .
We have , therefore , felt justified in marking them as maxima , indicating that are very large .
Canna and Portree are so disturbed that but little reliance can be placed upon observations taken there , but they both give maximum values of the Vertical Force ( Plate VIII ) .
At Corpach and Glenmorven we find that Mr. WELSH'S observation gave maximum values of the Vertical Force .
At two of the most northern stations the Horizontal Force is a minimum ; at three of the most southern it is a maximum ( Plate VIII They are marked with and respectively .
From this map we can form an opinion as to the magnetic constitution of the district .
Points of high Vertical Force , and centres to which the needle is attracted , cluster thickly along a line which passes from Elgin to Inverness , thence along the line of the Caledonian Canal to Corpach , and so to Mull .
Another imilar line runs north through Sky , and indications of a region of low Vertical Force , which separates the two , are given in tho minimum values found at Gairloch and Loch Inver .
Thers is a subsidiary centre of attraction near Strachur .
The view that a general attraction is exerted toward the centre of the district is supported ths occurrence of maximum and minimum values of the .
Horizontal Force in the south and north respectively .
The minima at Loch Boisdale and Soa and the maximum at Loch Maddy ( which necessarily follows ) indicate a strong subsidiary centre of attraction to the south of these places .
BALFOUR EWART , arguing from results , placed such a centre to the south of Mull , but as we shall show , it is probably to the west of that island ( ' Brit. Assoc. Report , ' 1859 , p. 190 ) .
It must be remembered that all these conclusions follow from the mere inspection of the results of the observation with no more calculation than is necessary to reduce them to the same epoch .
In the next map ( fig. 25 ) we show the results of the calculations carried as far as possible , i.e. , to the point of deducing the disturbing forces .
The shaded parts are the of positive ( i.e. , of great ) Vertical Force .
The boundaries are fixed by the very rough method of assuming that between neighbouring stations .
the rate of change of the disturbin force is uniform .
The arrows represent the Horizontal disturbing Forces in magnitude and direction .
The dotted arrows represent the disturbing forces calculated from Mr. WELSH'S survey .
There are only two or three discrepancies , and they all occur , as we anticipated , near vaHey lines .
The most remarkable is Stornoway .
There can be no doubt as to the .
A. W. RUCKER AND DR. .
E. THORPE ON A MAGXETIC relations uncertain , as the direction of the Horizontal Force would lie nearly along the valley line .
direction deduced from WELSH'S observation would place it in the next district .
The same remark applies , though more doubtfully , to Ardrishaig .
If , however , these three or four border stations be put aside , the map leads to a consistent view of the magnetic state of the district , which is in exact accord with that previously arrived at .
The ridge line lies in the region of greatest Vertical Force disturbance .
The Horizontal Forces tend towards that region , and at points within it are , on the whole , directed to the ridge line .
The discrepancies between our results and those of Mr. WELSH are hardly , if at all , greater than those between our own , when repeated .
It is remarkable that the south-western stations indicate a centre of attraction out at sea , but our results are completely confirmed by WELSH'S at Glen- morven .
We should certainly have expected priori that Mull , which is highly basaltic , would be a centre of attraction .
Between Elgin and Banff there is a strong local centre .
WELSH observed the Horizontal Force at Banff only , but the Declination obtained by him at Elgin is less than that at Banff , the difference being about 10 ' , as against 7 ' given by our survey .
The two sets of observations are thus in agreement . ?
he Scotch Coal-field District .
This district is bounded on the north by the valley line which forms the southern boundary of that which hasjust been discussed .
The southern boundary is inclined to the magnetic meridian at an ] which does not give any special advantage to either the Horizontal orce or the Declination as a means of determining its position .
We have therefore taken the Declination line corrected ( as described on p. 296 ) by WELSH'S observations at Makerstoun and Melrose .
There is not a single abnormal station in this district , and the results of WELSH'S observations fit in very well with ours .
At Air , indeed , there is a very considerable discrepancy , and the forces at Alford and Gretna are larger than we should have expected if they are regarded as the resultants of the attractions of the two regions , close to the mdaries of which these stations are situated .
All the four stations at which the Horizontal Force is a minimum , viz. ( taking them in order from the west ) , Cumbrae , Row , Stirling and Dundee , are situated close to the northern boundary of the Scotch Coal-field , in which there are large masses of basalt .
The ridge line runs through the middle of this basaltic district and passes through or near two regions in which the disturbance of the VerticaFForce is positive ( i.e. , in which the force is great ) to the east and west respectively .
At Cumbrae the Vertical Force is a maximum , and the existence of a region of high Vertical Force in this neighbourhood is remarkably confirmed by the fact that WELSH'S observations give a positive disturbance at Ardrossan a few miles further south , though at the neighbouring stations both to the south and north it is negative ( fig. 28 ) .
On the whole then MR. A. AND DR. T. E. THORPb ON A MAGNETIC Scotch isomagnetics are linear functions of the coordinates which determine the geographical position , and that the errors involved in the assumption would chiefly affect stations such as these , which are on the border of the district .
If it is necessary to prove that these results are independent of the method of calculation , Fig. 28 is sufficient for the purpose .
The lines of equal Horizontal Force converge in a most remarkable way towards North Wales , which is one of the most certain indications of a centre of attraction .
Again the Declinations at Holyhead and Llandudno differ only by , while the difference of longitude corresponds to 25 ' .
This again is strong vidence of the existence of a centre of attraction between them .
If , as appears probable , there are two regions of high Vertical Force in the district , it is possible that the northern one may be connected rather with the Cumberland Lakes than with North Wales .
However this may be , and while fully admitting that it requires further study , we think that this district appears to obe .
the rules which hold good elsewhere .
The North-Eastern District .
This district is bounded on the north and west by valley lines which have been already described .
On the south we have drawn a line which passes through the centres of two regions of negative Vertical Force disturbance .
The southern portion lies within the limits of our special surveys and has been so fully discussed that nothing need be added here .
The northern part presents a feature of peculiar interest .
The results as directly observed do not furnish such clear indic , tions as in a district where the presence of crystalline rocks is more obvious .
The isogonals are , however , bent in the peculiar manner which indicates centres of attraction .
Thus the line ( Fig. 29 ) points to such a centre to the south of Thirsk , the isogonal points to another between Hull and Gainsborough .
As both these lines depend in part upon the same stations , it is satisfactory to find that they are completely confirmed by the quite independent line which runs nearly due east and west between Lincoln and Mablethorpe .
The Vertical Force netico which corresponds to metric units bends southward to Thirsk in a way that is again an indication of a centre of attraction .
There are thus signs of an attractive region lying between the following pairs of stations , viz. , Mablethorpe and Lincoln , Hull and Gainsborough , Thirsk .
and Leeds .
On turning to the disturbances ( Plates IX .
and X we find that both the Declination and the Horizontal Force give nearly coincident ridge lines running along the line just indicated .
The declination line passes northward to the Cumberland Lakes and southward to the Wash. The Horizonal Force line turns south amid bhe Yorkshire Hills and runs toward North Wales .
MR. A. W. RUCKER AND DR. T. E. THORPE ON A MAGNETIC Vertical Force , and therefore the direction of the Horizontal orce cannot be determined from the rule .
The Wicklow and Arklow mountains , which are composed of granite , do not appear to exert any important effect on neighbouring stations , but there is a region of high Vertical Force in Wexford .
of the United Kingdom .
In Plate XIII .
we have attempted to represent the magnetic state of the whole country by bringing together the results of our studies of the magnetic districts into which it may be divided .
The valley and ridge lines are shown .
The unshaded parts are regions of negative Vertical Force disturbance , and the three shades employed indicate that the disturbance is greater than but less than 01 , greater than 01 but less than 02 , and greater than021netric unit respectively .
The lines representing the Horizontal Forces are drawn to a scale on which 1 mm. metric unit , except at Canna and Soa , where the disturbances are so large that the arrows would be inordinately long .
Considerations which have been adduced in the egoing discussion have led us to depart , in a few minor points , from the strict rules by which the valley and ridge lines have been drawn .
Thus , the line which separates the Highland and Scotch Coal-field Districts has been drawn so as to include Crieff in the latter .
The rather uncertain valley line in Mid-Wales , which was taken from the Horizontal Force disturbances ( PJate is replaced by the closely neighbouring ridge line , taken from the Declination disturbances ( Plate X This would necessitate valley lines on each side of it , the position of which is uncertain .
In Ireland too , the ridge line which runs from Antrim to the neighbourhood of Kells , is not very definite , and it is best to consider the clearly marked portion of it in Antrim as only doubtful]y connected with the centre of attraction near Kells .
These points must be left for future investigation , but the unquestionable existence of widespread regional disturbance in the districts we have specially studied in England , together with the general agreement between the two maps of Scotland , deduced from the surveys of 1857 and 1886 , leads us to hope that Plate XIII .
gives the first approximation to a map of the disturbing magnetic forces in play over the whole kingdom .
On the between the Magnetic Geological Constitution of the Magnetic Districts .
Up to the present , we have discussed the various districts into which we have divided the country from the point of view of their magnetic peculiarities only .
SURVEY OF THE BRITISH ISLES FOR THE EPOCH JANUARY 1 , It now remains to investigate the question whether any connection can be established between these and their geological characteristics .
It is well known that certain varieties of crystalline rocks are often more less magnetic , and that when they are permanently magnetised the poles are sometimes very irregularly distributed .
We have , however , thought worth while to investigate the magnetic state of some pieces of diorite and basalt , brought from Malvern and Canna respectively .
The fragments were cut into small rectangular blocks , delicately suspended and tested .
We have to thank Dr. HOFFERT , Demonstrator in the Physical Laboratory of the Science Schools at South Kensington , for undertaking this part of the work .
The observations were very tedious , and were , for the most part , made by Messrs. GRAy , ANDERSON , and WILKINSON , students in the laboratory .
Of several blocks from Malvern only one showed any polarity in its natural state , but when placed between the poles of an electromagnet it became magnetised by induction , so that the time of oscillation was reduced from to .
The stone brought from Canna was part of a basaltic column and its upper , lower , east and west ends were marked .
It showed polarity , but the upper end was a north-seeking pole , so that it was magnetised in a direction opposed to that which would be induced by the magnetic field of the earth .
The moment due to the permanent magnet sm was calculated by three different methods , .
: ( 1 ) by the difference of time of oscillation when the direction of the field ( about twelve times as strong as that of the earth ) was reversed , ( 2 ) by the deflection when the stone was placed E. and W. , and ( 3 ) by the difference of times when the position of the stone was reversed .
The results obtained vere 0027 and .
unit .
The periods of oscillation were and in the earth 's field and the artificial field above described .
Tho general conclusion arrived at was that , as the volume was about 1 , the permanent intensity of magnetisation about C.G.bi .
unit , and that in a field of strength the induced intensity was about F. We have examined the relations between the magnetic disturbing forces and the geology of the area of the survey by means of a geological map which has been specially prepared for us under the kind superintendence ofProfessor JUDD ( Plate Details are disregarded , but the principal masses of basaltic and non-basaltic crystalline rocks and the main groups of the sedimentary formations are clearly distinguished from each other .
This may be compared with Plate XIII .
, in which the magnet disturbin forces and the ridge and valley lines ate shown .
As the ridge lines are drawn according to an arbitrary rule , they are only intended to draw attention to the districts in which loci of attraction probably exist .
We have no real knowledge of their distances from the stations between which they run .
In the geological map , therefore , the stations on each of the principal ridge lines ave been connected by lines , the spaces ) closed have been shaded , and thus the within hich the main loci of attraction probably lie are clearly indicated .
It must be remembered that the outlines of these districts depend largely on the accidental positions of stations which were f ; elected without reference to them , and that we must rather expect such rough indications of relations between magnetic and geological facts as may serve to guide future investigations than complete and unmistakeable harmony .
When regarded from this point of view , however , we think the results are very suggestive .
In the Highlands one region of attraction encloses the Caledonian Canal , another is evidently in close relation with the basaltic masses in Sky , Glenmorven , and Mull , though , as has been pointed out , the main centre of attraction appears to be to the west of these islands .
A third region in Scotland encloses the basaltic rocks of Arran and of the Scotch Coal-fields .
The fact that the Fifeshire basalt lies outside it is probably due only to the accidental circumstance that we have no stations between Stirling ( 47 ) and Dundee ( 21 ) .
A fourth region evidently consists of the Antrim basal .
There is a fifth in Connemara , where the rocks are granite .
With regard to North Wales , if we consider it as formin one district with shire , we see that a line drawn through the centre of the basaltic rocks would first run from east to west , hen nearly north , and finally turn to the west between Anglesea and Carnarvonshire .
The district is thus very irregular and the ridge lines do not give much information , but we must poin out that Horizontal Forces at the stations which border it all tend inwards the axis above suggested .
This is true of Holyhead ( 90 ) , Llandudno ( 106 ) , Llangollen ( 107 ) , Shrewsbury ( 138 ) , Aberystwith ( 55 ) , and Pwllheli ( 128 ) .
If this suggested relation is hereafter verified , every considerable mass of basaltic rock in the kingdom will be closely connected with a region of magnetic attraction .
Of smaller masses it is to be noted that at ] moutb ( the disturbing force acts southward toward the serpentine of the Lizard , and that the elatively small masses in Pembrokeshire and Wexford are within another region of attraction .
Our measurements do not sign any particular importance to the outerop of basaltic rock near Limerick ( 185 ) nor to the dykes in the north of England .
It is , however , curious that a line drawn through Melton Mowbray and the Wash peak ( indicated by circles ) passes towards Wales through the only basalt in the Midlands .
Taking the evidence as a whole , we think we are justified in saying that large masses of basaltic rock indicate regions of magn tic attraction .
The other crystalline rocks appear to be much less important magnetically .
Thus the Malverns , though a strong local centre , do not disturb a district of any considerable magnitude .
If the effects of the two classes of rocks when they appear on the SURVEY THE BRITI@H ISLES FOR THE EPOCH JANUARY 1 , 1886 .
813 surface are in such marked contrast , may be open to question whether strong ..magnetic attraction in a district in which no rocks appear on the surface does not indicate not only crystalline but basic crystalhne rocks beneath it .
However this may be , there are two regions of attraction which not connected with basaltic rocks on the surface , at all events as the main cause of their peculiarities .
One of these runs westward from London io the South Wales Coal-field , and the directions of the disturbing forces in Wexford are such as would be caused if it crossed the Irish Channel .
This extension is doubtfu ] , but the line is most clearly marked ight across England , and its general direction is such as to make it almost impossible to avoid the conclusion that it is connected with the palaeozoic ridge , the existence of which , long ago predicted by Mr. GODWIN AUSTEN , has been proved by deep borings near London , and which is supposed to connect the Welsh and Belgian Coal-fields .
If the palaeozoic rocks are nearer the surface here than elsewhere , the crystalline rocks may approach it also within a moderate distance , and if they are susceptible to magnetisation the observed results would follow .
The palaeozoic rocks are supposed to form basins , and if those beneath them have a similar outline , it would be possible to explain a centre of attraction such as the Reading peak .
The last region of attraction runs the Wash through south-east Yorkshire toward the Cumberland Lakes , and , as Professor JUDD first pointed out to us , it includes the line to the north of the Humber , along which the oolitic and liassic strata thin out rapidly , and where , therefore , the crystalline rocks are probably suddenly brought much nearer to the surface .
It is continued northward ) ward the igneous masses in Cumberland .
The centre of attraction which apparently exists at Kells , in Ireland , is not sufficientJy accounted for by the surface geology of the district in which it is placed , and we have not felt ourselves justified in enting it as connected with the Antrim basalt , as the fact that this connection requires further confirmation has been already pointed out .
On the whole , then , we think we may assert that every region of magnetic attraction , with the possible exception of that near , is marked either by the presence of basalt , or by some geological peculiarity which makes it ] or even probable that within it crystalline rocks , capable of affecting the magnet , are nearer the surface than elsewhere .
This is possible as regards the line of fault marked by the Caledonian Canal , and probable as regards south-east Yorkshire and the south of England .
The former district is remarkable from the fact it was far more strongly affected by the earthquake of Lisbon than the rest of the British Isles , and this may , perhaps , indicate that it has special relations to the primary rocks , which wouJd account for its magnetic importance .
MR. A. W. RUCKER AND DR. T. .
THORPE ON A MAGNETIC On comparing the geological map with Plate XIII .
, on which the disturbing are shown for the whole country , a difficulty arises from the fact that the Forces appear , on the whole , to be greater over the districts in the east south of England , where the later sedimentary rocks occur , than over Wales Ireland , where , from their absence , it would primd facie appear probable that downward attraction would be greater .
Of course many hypothetical explanations could be offered of the fact , such as that the primary rocks in England might possibly contain larger quantities of ferruginous matter , \amp ; c. , but we must be content with observing ) if a fairly uniform increase in the disturbing Vertical Force were to take place from east to west , it is very doubtful whether we should have detected it .
Probably it would cause a disturbance of the first order , the terrestrial lines would be deflected , and the disturbances at distant points would not be comparable .
If the true Vertical Force isomagnetics could be prolonged beyond Wales into another district in which the tertiary strata re-appeared , a southward trend in Wales would indicate an increase in the force .
On looking at Plate VIII .
we think it win be admitted that the slope of the lines of equal Vertical Force is greater in Wales than in England , which is in harmony with the existence of a southward bend , but the fact that they are extremely irregular and terminate on the west in the sea makes any certain deduction impossible .
We must , therefore , regard the distribution of the Vertical Force disturbances as presenting some difficulty , and must empbasise the necessity of using them only to compare neighbouring stations .
The Causes of and Regional gnetic Forces .
It has long been known that distortions of the isomagnetics occur chiefly in the neighbourhood of crystalline rocks , and it has been generally assumed that this is due to so-called rock magnetism , the rocks being magnetised either permanently or by the inductive action of the earth 's field .
Dr. NAUMANN , in the work already quoted ( ' Die Erscheinungen des Erdmagnetismus , ' Stuttgart , 1887 ) , has recently opposed view .
The arguments which may be brought against it are : ( 1 ) That rocks brought from considerable depths do not exhibit magnetic qualities until they have been for some time upon the surface ; ( 2 ) That the effects which rocks or mountains produce on the magnet , even if very great when the distance is small , diminish so rapidly as the distance increases that they are quite insufficient to account for the widespread effects which are attributed to them ; ( 3 ) That extensive local magnetic disturbance is associated rather with geological faults than with the presence of igneous rocks ; ( 4 ) That the cause of the phenomena is to be looked for in the effects produced on SURVEY OF THE BRITISH ISLES FOR THE EPOCH UARY 1 , 1886 .
315 earth-currents by dislocations of the rather than in rock magnetism .
We will consider these arguments in the inverse order .
to that in which we have stated them .
Mr. PREECE , F.R.S. , Chief Electrician to the General Post Office , has been good enough to have measurements of the earth-currents made in several of the districts in which we have found large disturbances of the Declination .
The currents flowing between various post-offices have been observed , and the directions and intensities of the currents noted .
The following Table gives the data obtained on telegraph lines between Melton Mowbray and stations in its neighbourhood .
The letters P.D. signify Potential Difference .
These potential differences per mile are much smalJer than those which occur during magnetic storms , and which on the earth-current theory must , we suppose , be regarded as the causes of the deflections of the Declination needle which then take place .
During a very violent storm in 1881 Mr. PREECE found a P.D. of volt per We will take for comparison a less extreme case .
Yearly records of the movements of the Declination needle , and of the registers of the earth-currents apparatus , are Qhed by the Greenwich Observatory , and the Astronomer-Royal informs us that half an inch on the earth-current registers corresponds approxlmately to a P.D. of one volt in circuit .
On September 10 , 1886 , an iucrease of 16 ' took place in the Declination between and , and between and a change in the intensit , v of the current in one of the circuits occurred which corresponded to a change of P.D. of about volt .
The direction of the current was such as to produce the observed movement of the needle , and if wo regard the current as its cause , since the distance between the earth plates is 3 miles , we find that a deflection of 16 ' was produced by a potential difference of volt per mile .
There are two earth-current circuits mutually at right augles , both of which are inclined at about to the magnetic meridian .
The second was but slightly affected , and , therefore , the effective difference of the potential was about volt per mile nearly .
At Melton Mowbray we find the disturbance of the Declination to be 33 ' W. at one MR. A. W. RtTCKER AND DR. T. E. THORPE ON A MAGNETIC of the two stations .
It changes very rapidly , falling to 26 ' W. within a mile quarter , and to nearly.30 ' E. at Loughborough , which is thirteen miles distant .
currents produce it they must therefore be very local .
On the other hand we know that the disturbances produced during magnetic storms are simultaneous over areas such as that of the United Kingdom , and if the currents produce the deflections , also must be widespread , so that if we assume that the same earth-current is required at Melton to produce a permanent deflection as is observed at Greenwich when an equal temporary deflection takes place , we are not overstating the case .
Now , during the whole of the year 1886 , there were only a few occasions on which the earth-currents were stronger than on that which we have selected as an and we may therefore say that on the earth-current theory there should be permanently at Melton a potential difference of volt per mile ( for the smaller deflection of 26 ' is times that observed at Greenwich on the selected occasion ) an amount which is only registered at Greenwich during violent storms , and is nearly thirty times greater than that ) served in a circuit only two miles in length in the neighbourhood of Melton itself .
So far we have dealt only with the magnitudes of the currents .
The case becomes much stronger when we consider their directions .
On all the circuits out of Melton Mowbray referred to in the table on p. 315 , the earth-currents flowed from that station .
Hence , on all the currents except that to Oakham , the direction of the current was such as to produce a deflection of the North Pole to the east , i.e. , in the opposite direction to that which was actually observed , while the P.D. between Oakham nd Melton was the smallest of those measured , being only volt per mile .
We do not lay stress on this , as the distance was perhaps rather too great , the main fact being that the earth-currents between Asfordby and Melton , on the same side of the latter town as that on which our second station was placed , were not only much too small ( if we mayjudge from what is observed at Greenwich ) to produce the observed deflection , but that they were actually in the wrong direction .
To make certain that nothing in this argument depends upon the particular deflection obtained at Greenwich on the selected occasion , we have taken enty other examples in which Declination disturbances of from 3 ' to 22 ' were accompanied by changes in the earth-currents .
To avoid the necessity for correcting for diurnal variation they were chosen from the nocturnal hours when the magnet is steady , and occasions were selected on which one circuit only was appreciably affected .
Except in these particulars they were chosen haphazard .
If we divide the E.M.F. per mile in volts ( V ) by the Declination disturbance we get a number which expresses a relation between the two , and which , if they were cause and effect , ought to be constant .
The following table shows that the selected example was not particularly favourable to our views .
As the first station at Melton was on the south side of the town we assume that the potential difference was the same as that between Melton and kham .
At the second station we take it to SURVEY OF THE BRITISH ISLES FOR THE JANUARY 1 , 1886 .
317 be that between Melton and Asfordby .
Tn both cases the whole is supposed to be effective , i.e. , it is not resolved along the magnetic meridian .
A negative sign means that the current flowed in the direction opposite to that which would be required to produce the observed deflection .
We are quite aware that an argument of this sort is open to criticism .
The exact relation between earth-currents and magnetic storms is uncertain ; in short circuits the earth-currents may be masked by those due to the earth-plate , and we have been compelled to assume that the conditions at Greenwich and Melton are the same .
We should not , therefore , have put the argument forward had conclusion depended on any nice balance of figures .
As it is , we think it supports the view that the perinanent earth-currents at Melton Mowbray are very much less , and less extended , than the temporary currents observed at Greenwich during ordinary but considerable storms , though the permanent Declination disturbance which they are supposed to produce is of the same order as the temporary deflections which are observed simultaneously with the currents at Greenwich .
Similar results were obtained near to the Reading and to the Wash turbances .
If these are produced by earth-currents they must circulate round the peaks , and thus the potentials at points on opposite sides of them should be different .
Near the Wash the direction of the telegraph wires is not very favourable for a but we owe to the kindness of Mr. PREECE , to whom we must again express our obligation , series of measurements made at Reading and Windsor .
Of these it is only necessary to say that , though Reading and Windsor are on opposite sides of the focus of disturbance , and though the needle is deflected to the west at Windsor and 6 ' to the east at Reading , so that the assumed current circulating round the peak ought to run in opposite directions through them , the observed earth-currents were so small that no measure of their nagnitude could be Laken .
It might be urged , in answer to these arguments , that the earth-currents by which local disturbances are produced are not mere surface currents , but that they flow through the mass of the earth , possibly where the strata are ( as recently suggested by Professors LAMB and SCHUSTER ) of higher conductivity .
To this it may be replied that the greater the depth at which the currents are supposed to be situated , the more difficult is it to account for the fact that their effects are so local .
A more conclusive answer is the fact that our observations show clearly that ths needle is attracted to certain lines or points round which currents would have to circulate in the same direction with hands watch ) in order to account for the facts .
If the magnetic field of the earth is due to currents flowing from east to west , it is easy to that they might be deflected by layers of more or less than average conductivity ; but why should they form local eddies , in which the flow is always in the same direction ?
They must circulate similarly round the Readin and Wash peaks , round Kells , probably also round the Antrim basalt and the Toad-stones of Derbyshire .
In accordance with Commander CREAK'S investigation , they must flow round magnetic islands in opposite directions in the two hemispheres .
They must flow in different directions on the two sides of the palaeozoic ridge in the southern counties , of the Malverns , of the Yorkshire ridge , of the Scotch coal-field , and of the Caledonian Canal .
It may be possible to imagine physical causes which would account for such a state of things , but it does not appear easy to frame an hypothesis which shall be more probable than that involved in the theory of magnetic rocks .
As regards Dr. NAUMANN'S view , that geological faults determine local magnetic action , it is disputed in its application to Japan by Dr. KNOTT , who has superintended the recent magnetic survey of that country .
We find at a mass of crystal ] in lock bounded by a fault .
Either the rock or the fault ntay be supposed to be the cause of the attraction toward the ridge which is undoubtedly exerted on the needle ; but whereas the rock is susceptible of magnetisation , and its effect on the magnet is precisely such as would be produced if it were magnetised by induction in the earth 's field , there is no shred of direct evidence to prove that the fault is capable of causin or deflecting earth-currents , so as to make tbem flow in opposite directions on its opposite sides .
It more probable that the action of faults is due to the displacement of crystalline rocks in their neighbourhood ; and that a fault is , at all events , not a necessary cause of regional attractions is proved by our observations in Scotland .
Two great fault lines traverse that country , one coinciding with the Caledonian Canal and the other running from Stonehaven to the mouth of the Clyde .
The first of these is , and the second is not , associated with a locus of attraction .
Taking the next argument in order , we agree that mountains which exercise considerable influence on the magnet when close to them produce no effect at distances which are small relatively to the range of regional disturbances .
It must , however , be remembered that the depth from the surface to magnetic rocks concealed by overlying strata may not exceed the horizontal distances at which the Malverns and the Wash affect the Declination , and that their influence on the Vertical Forces might extend over vast areas .
If , as would often be the case , they were not horizontal , but approached the surface by a gentle slope , the magnet would certainly tend to turn SURVEY OF THE BRITISH ISLES FOR THE EPOCH JANUARY 1 , 1886 .
319 towards or away from the nearer and more elevated portions , according to the nature of their magnetisation .
If the slope were less than the angle of Dip , the sides of the sub-terrestrial hill would be magnetised , so as to attract the north-seeking pole , and the empirical rule that the Horizontal Forces tend toward places of greatest Vertical Force would be amply and completely explained .
Dr. argument loses a great part of its force if we regard a widespread disturbance of the isogonals around a mountain range as due not to the direct action of the mountains but to a far reaching mass of rocks of which they are the culminating point .
Lastly , as to the statement that rocks containin ron only become magnetic when brought to the surface\mdash ; without impugDing the actual observations , we can only say that so important a generalisation requires a much more basis of fact than any that is provided for it .
If the vern granites produce in their immediate neighbourhood a magnetic effect , it is evident that rocks , the susceptibility of which can only be detected by refined methods , may in large masses deflect the lines of force in the earth 's magnetic field to an appreciable extent .
On the whole then , while fully admitting that there is room for much further investigation on this head , and that any view is more or less hypothetical , we do not think that any theory has hitherto been proposed which offers fewer difficulties than of rock magnetism .
If the cause of the magnetic disturbance is induced magnetism , it would vary with the direction and intensity of the earth 's magnetic field , but the change could hardly be great relatively to the secular variations of the elements .
It is therefore important to note that in the 'Phil .
Trans. ' for 1870 ( vol. 160 , p. 274 ) , Sir E. SABINE gives the Declination at Greenwich and Kew as having been practically identical in His values Greenwich Kew As according to his isogonaIs the Declination at Kew should , at that epoch have been 9 ' greater than that at Greenwich , whereas , according to his table , it was 2 ' less , the difference due to disturbance was ( i.e. , the Kew value was too small ) , whereas in 1860 it was , in , and is now 10 ' ( p. 270 ) .
If then we accept Sir E. table as correct , we must assume that a disturbance difference amounting in all to 22 ' was established between and 1860 , which has remained constant during the twent.-nine years which have elapsed since the latter epoch .
We , therefore , wrote to Mr. WHIPPLE asking him for information as to the observations given by SABINE , as having been made Kew in .
It should be noted that the authority quoted by SABINE is " " which indicates an official origin for the value given .
MR. A. W. RUCKER AND DR. T. E. THORPE ON A MAGNETIC Mr. WHIPPLE writes , " " I have now looked over the MSS .
papers I can find , referring to this Observatory in 1842 , and I am quite unable to find any reference to the observation quoted in the 'Phil .
Trans. ' for 1870 , p. 274 .
There was no here at the date , nor any official publication , hence I do not see how " " Observator or , as printed , " " could be any authority .
Again , in SABINE'S contribution , 'Phil .
Trans 1.849 , p. 208 , vich and Bushey are both given with the names of the .
obseryers , and I think there can be no doubt that if SABINE had observed here in 1842 , he would have certainly quoted the observations in that table .
I fear we must consider it somewhat too hypothetical to trust to On the whole then , it appears to us most likely that the number given by SABINE .
was not deduced from an observation made in 1842 , but was reduced to that epoch from the Kew Observatory records , which did not begin till some years afterwards , and that in the deduction some numerical blunder was made .
SUMMARY .
We may now attempt to sum up the result of our studies of local and magnetic forces in the United Kingdom .
We have proved beyond possibility of doubt that two centres of attraction exist near Reading and the Wash respectively , and in the former case it is evident that the Horizontal disturbing Forces tend towards regions of maximum Vertical disturbing Force .
We have then gone through the United Kingdom district by district , and have shown that the same rule holds good everywhere .
Putting aside stations of maximum or minimum Vertical Force and stations on valley lines ( which do not furnish real exceptions ) there are not half a dozen which are clearly anomalous in tho whole 200 ; nor has this result been obtained by splitting the whole area of the survey up into a large number of small districts which have been carved cut so as to secure an appearance of uniformity of behaviour .
omitting the border stat , ions on the south of Scotland , in Devonshire and the Channel Isles and Dover , the magnetic constitution of the rest of Great Britain is accounted for by five principal ridge lines only .
The first is coincident for a great part of its length with the direction of the great fault of the Caledonian Canal .
The second and third are connected with the masses of basalt in the Western Isles and the Scotch Coal-field respectively .
The fourth runs for 100 miles parallel to , if not coincident with , a line in Yorkshire and Cumberland along which geologists believe that crystalline rocks may occur near the surface , and sends out a branch toward the igneous rocks which occur in North Wales .
The fifth runs for nearly 200 from London to Milford Haven along another similar geological line , and sends out branch to the south coast from near Reading .
SURVEY OF THE BRITISH ISLES FOR THE EPOCH JANUARY 1 , 1886 .
321 It is probable that to these a sixth in North Wales and Shropshire should be added .
There are indications of other minor centres at Malvern , in Derbyshire , and in the neighbourhood of Charnwood Forest , but the number of stations involved is small .
The vast majority are included in the few and simple systems above described .
These results are not the outcome of the calculations only , for all principal conclusions can be drawn from the observations alone .
Mr. WELSH'S survey , though including fewer stations at which all the elements were determined than ours , though omitting almost altogether stations in the Western Isles , though worked up ( as regards the Dip and Declination ) by means of a different system of geographical coordinates , and in the case of all three elements by means of an assumption as to the linearity of the isomagnetics which we have abandoned , confirms our conclusions .
In four only of the 28 of his stations which fall within our districts , viz. , Stornoway , Pitlochrie , Air , and Dumfries is there an important difference between us as to the direction of the disturbing forces .
In the most highly disturbed istricts , at Loch Inver , Kyle Akin , Broadford , Glenmorven we agree as to the order of the magnitude and as to the direction of the forces .
At Cumbrae and Fairlie we find closely neighbouring stations of positive surrounded by others of negative Vertical Force disturbance , and WELSH confirms us by a similar result at Ardrossan , not ten miles off .
The , se coincidences between the results of the two surveys , and between the magnetic peculiarities and the geological constitution of districts cannot be accidental , and we venture to assert that , throughout the kingdom , the lines of Horizontal magnetic disturbing Force tend towards regions of maximum Vertical disturbing Force , which are themselves defined by the presence of crystalline rocks , and especially of basalt , either visible on the surface or concealed by superimposed masses of sedimentary strata .
Beyond this general conclusion we do not wish at present to go .
The detailed constitution of our principal magnetic districts ( except in the case of the Reading and Wash disturbances ) has yet to be investigated .
We have not discriminated between various possible causes of a decrease in the Vertical Force , such as the removal of the atbracting mass to a greater depth , or the formation of repulsive poles , either by induction or permanent magnetism .
We do not think the last word has been said on the cause or causes of local and regional forces .
All these require and will , in the future , no doubt , receive attention .
We trust , however , that by following out to their logical conclusion the premises of a not improbable working hypothesis , we have succeeded in showing that local and regional forces obey certain simple laws , and that by means of these the kingdom can be divided into magnetic districts in which the relations between the direction of the disturbing forces and the main geological characteristics are so suggestive as to be worthy of careful statement and further investigation . .
A. W. RUCKER AND DR. T. E. TFtORPE ON A MAGNETIC TABLES F0R CALCULATION 0F THE MAGNETIC ELEMENTS 1N THE FUTURE .
We conclude this paper with Tables , by means of which the of the magnetic elements may be determined during the next few years for any place in the United Kingdom .
For this purpose we should know as accurately as possible\mdash ; ( l ) the values of each element at the station at the epoch of the survey ; ( 2 ) the secular change ; ( 3 ) the local disturbance .
To determine the first of these , we give the next three Tables ( VIII .
, IX .
, X in which the values of the elements deduced from the general formulae are given for all intersections within the kingdom of lines of titude and longitude which correspond to whole degrees , together with the variation per degree of latitude and longitude .
The method of using them requires no explanation .
The values in the Table of Horizontal Forces display some irregularities .
This is partly due to the fact , that a discontinuity is introduced along the line , on opposite sides of which we have used different formulae .
The formula also , which is applied to the southern district , does not represent parallel lines , but lines of which the slope is a periodic function of the latitude .
We have slightly smoothed the irregularities , a process which will introduce some discrepancies between the numbers given by the Table and by the formulae , but we have thought it better not to attempt to do away with all traces of the discontinuity .
SURVEY OF THE BRITISH ISLES FOR THE EPOCH JANUARY 1 , 1886 .
323 The determination of the rate of secular change is a more difficult matter , but at the completion of our work we have more data at our disposal than in the earlier stages , when coefficients had to be chosen for the reduction of observations .
The plans adopted in the case of each element are as follows:\mdash ; 1 .
We have deduced from Sir F. EVANS'S Map for .
cit. the Declinations at 24 points distributed uniformly all over the kingdom , and have compared them with the values given by our Table VIII .
, on .
The secular corrections thus aalculated show an increase from east to west amounting to about and per degree of longitude in latitudes and respectively .
There is also an increase with latitude above the latitude The results are exhibited in ttJe following Table:\mdash ; TABLE XI.\mdash ; Mean Secular Change of Declination per between 1872 and 1886 .
This is in fair accord with the observations at individual stations given on p. 88 , but as Sir FREDERICK'S isomagnetics were not deduced by any definite system of calculation , but little importance must be attached to minor variations and discrepancies .
We do not , for instance , feel justified in assuming that there is a real increase of the secular coefficient to the south of lat. , especially as M. MOUREAUX'S values in France for corresponding intervals are less than ours .
It is also in accord with the results of the comparison of our observations with those of WELSH , which show that between 1857-1886 the secular change in increases with the latitude and with west longitude .
We think , therefore , that the most that can be said present is that the secular change at the time of our survey , as given by the mean of the Greenwich and Kew values ( p. 91 ) , was in the neighbourhood of London about per annum , and that a comparison with Sir F. EVANS'S results sbows that it is about greater in the south-west of Ireland , and about 2 ' larger in the north-east of Scotland .
Hence by smoothing the irregularities in Table XI .
above , and reducing all the numbers so as to give the present rate at London we get the figures in Table XII .
: p. 825 .
These may be checked in future by means of the observations at Kew , Greenwich , Falmouth , , and Valentia , but we do not think a comparison of two observatories only leads to trustworthy laws of rate of change of secular variation with geographical position .
There can be no question that Declination change on the whole increases with the latitude , yet the secular coefficient is less at present at Stonyhurst than at Greenwich and Kew .
It is therefore very much to be regretted that the observations so sedulously carried on by the late Provost LLOYD , at Trinity College , Dublin , have oflate years been interrupted .
We may point out that the best position for a magnetic observatory would be in the centre of a widespread region of low Vertical Force .
The rocks would probably produce less effect in such a situation , and all the phenomena might be expected to be more normal .
Unfortunately Kew and Greenwich are within the range of the Reading disturbance , and Stonyhurst is in a region where the Vertical Force changes very rapidly both to the east and west of that place .
2 .
Inclination .
We have also in the case of the Dip found the secular change between and 1886 by comparing Table X , with Sir E. SABINE'S map for the earlier epoch .
The result confirms the conclusions previously arrived at , viz. , that the secular change diminishes with latitude and increases with west longitude .
As , however , the north-westerly are on the line of minimum change they probably do not form real exceptions to the general rule .
Partly } ) this method and partly by the map of assumed values ( p. 85 ) we have selected those in Table XII .
3 .
Horizontal Force .
This element when treated in the same way gives since a greater rate of secular change in the south than in the north , and in the west than in the east .
The latter conclusion is in accord with that of M. MOUREAUX .
We are now inclined to think that the high value of the secular change between the years at Greenwich led us to assume rather too high a value for the south of England .
The mean annual change for Greenwich between the years 1883-87 is 0020 and for .
The values for Greenwich in 1857 and 1887 are and , but in 1861 a new instrument was introduced , and to make the two values comparable we must subtract from the first .
This leads to a difference of or a secular change of Though this is a little greater than the present rate , we do not think the evidence is sufficient tojustify us in altering the rates obtained for Scotland by a comparison for the interval 1857-87 , and we assume that they are valid at the present time .
For the neighbourhood of London we take , and by the aid of SABINE'S and WELSH'S papers we hay :rrived at the values given in the following Table:\mdash ; The local disturbance at any place may be estimated from the disturbances at neighbouring stations as given in Plates IX .
, X. , and XII .
It is evident that this correction must be somewhat uncertain , but maps will at all events give information as to whether the disturbance is likely to be large or small .
As an example of the use of the tables and , we calculate the Declination for N. and long .
1o 30 ' W. on July 1889 .
From Table VIIT .
it was on January 1 , 1886 .
Fron Table XII .
the secular change is which reduces the value to From Plate IX .
we see that the station is in a region of negative disturbance , and that the true Declination is probably less than the calculated amount ] or 20 ' .
It is obvious that for this purpose much rougher methods of calculation would suffice , but the main reason for making the process as accurate as possible is that the values of disturbing forces can only be determined if the rates of secular change are carefully ussed and accurately known .
As we hope that these will be further investigated , we give the fullest data at our disposal for the calculation of the undisturbed values of ) elements .

