The object of this publication being to present in their most salient point of view the irregular movements of the needle and variations of the magnetic force, it is necessary in the first instance to ascertain and subduct from the observed changes everything of a regular periodical nature which the actual amount of our present knowledge has enabled us to ascertain. This has been accordingly done, so far as Toronto and Van Diemen's Land are concerned, both for the diurnal, monthly, and annual fluctuations of the magnetic elements as far as it has been practicable yet to deduce them, and the results have afforded room for preliminary conclusions of no small interest, which have been stated by Colonel Sabine in his preface, and of which the following is a brief outline.

At Toronto the regular diurnal movement in declination does not consist in a simple uninterrupted progress and regress of the needle. Commencing from 2h P.M. its movement is continuous to the eastward till 10 P.M., it then returns westward (through a comparatively small angle) until 2h A.M., when its eastward movement is resumed and continued till 8h A.M., after which its return is continuous to the west until 2h P.M. This second eastward progression is more decided in summer than in winter, and the total range of diurnal fluctuation is also more considerable.

At Van Diemen's Land (a station it is to be borne in mind almost antipodal to Toronto), the course of the diurnal oscillation corresponds with that above stated in all but one essential feature, viz. that the hours (in mean time at the station) of easterly movement of the north end of the bar at the one station are those of its westerly movement at the other, that the diurnal range being nearly the same in both, with a similar inequality in its summer and winter amount; a similar alternate progression and recess also prevails, and at the same hours.

These are certainly very remarkable features, showing a regular connexion between two stations so remote, carried out into what may be regarded as minute particulars. Falling in however with the generally received impression of the universality of the causes (whatever they may be) which produce the periodical fluctuations of the magnetic elements, they can only be regarded as contributions to our knowledge of details. It is otherwise with the results deduced by a comparison with each other of the observations recorded in this volume, not only at these two stations but also at St. Helena, and with those made by M. Kreil at Prague, as respects cases of unusual magnetic disturbance which occur (so far as we can yet perceive) casually, or at least non-periodically. Such comparison has enabled us at length, unequivocally, to state it as a general proposition, that the whole magnetic system of our globe is affected in the majority of cases of great disturbance. For it is found that if a list of days of great disturbance, independently noticed as such, and marked by extra observations on each station, be made out, these lists will be found to coincide in at least a majority of days, and more especially on those days when the recorded disturbances have been greatest. Of twenty-nine principal disturbances recorded in Colonel Sabine's Synoptic Table, some confined to a single day, others running through two or three successive days, and comprehending altogether forty-nine days, by far the greater part are shown to have manifested themselves at Toronto, Van Diemen's Land, and Prague, and fifteen are marked by extra observations at St. Helena. But though it is thus rendered certain that the whole globe is affected in many and great "magnetic storms," it is equally shown that the minute identity of particular shocks, which seemed to result from the earlier observations of this nature in Europe, cannot be maintained (as a general proposition) as traceable on anything like so extensive a scale.

Not the least interesting part of this volume consists in the notices at Toronto of auroral phænomena accompanying the extraordinary magnetic disturbances. They are many and remarkable, and can hardly fail to throw great light on this branch of the general subject.

This is not the proper place for theory, nor is anything more than an analogical illustration intended, if we compare the affections in question to what might be supposed to occur if we conceived the earth surrounded, besides the ocean and the air, with an electric atmosphere of excessive elasticity and mobility, in which were propagated from origins unknown to us, undulatory movements of every order, from the most minute local oscillations to waves affecting (almost in an instant, or in very short intervals of time, but varying in depth and amplitude with the geographical coordinates) its whole extent. Could such electric waves* be conceived as affecting the magnet, we might, form some idea of the mode in which particular shocks thin off as it were by distance of place, and are replaced by others of different local origin.

The difficult subject of the determination of the earth's magnetic force in absolute measure has been subjected to a further investigation by Dr. Lloyd. The difficulty, which is of a practical rather than a theoretical nature, arises from this, that the expression for the tangent of the angle of deflection of one magnet by another being expressed approximately by two terms of a series according to descending powers of their mutual distance, viz. the inverse cube and fifth power, with unknown coefficients, these have to be determined in Gauss's method by observations of deflection at two different distances, and by eliminations, in which process serious errors are introduced in the result by small errors in the observations. The object of Dr. Lloyd's present paper is to point out a means by which the quantity sought may be obtained without elimination, by observations at one distance only, thus diininishing both the trouble of the observation and increasing the accuracy of the result. This method depends on the assumption of an empirical law in the distribution of free magnetism in a magnetised bar inferred by Biot from Coulomb's researches, in virtue of which a simple ratio, dependent only on the lengths of the two magnets, subsists between the coefficients of the inverse powers above mentioned,-a ratio such, that on a certain simple assumption of the proportional lengths, the term depending on the inverse fifth power may be made to vanish ipso facto, and thereby get rid of the whole difficulty. Dr. Lloyd adduces several experiments confirmatory of these results.

The Annalen für Meteorologie, Erdmagnetismus und verwandte Gegenstände,' published by M. Lamont with the assistance of Messrs. Grünert, Koller, Kreil, Lamont, Pleiniger, Quetelet, and Stieffel, for the year 1842, is completed, and will be followed up by similar series in quarterly parts. In this collection are contained a multitude of important contributions to these subjects from all quarters, and more particularly magnetic observations from Munich by M. Lamont; and meteorological registers from Marseilles by M. Benjamin Valz, from Schlösse by M. Bayer, from Dorpat by M. Mädler, from various stations in Labrador and Greenland, from Utrecht by M. Van Rees, from Munich by M. Leonhardt, and a series of comparative observations from Stuttgard, Giessen, Carlsruhe, Vienna, and Parma, in which the barometric and thermometric observations are not stated absolutely, but only their dif*It is by no means necessary in this way of conceiving the subject, to assume an atmosphere of pure electricity (of which we can form no conception). But we may, for hypothesis sake, admit the existence of an atmosphere of some medium very much more rare and elastic than air, by whose compressions and dilatations electricity may be momentarily developed and absorbed, as caloric is by those of air in the phænomena of sound, manifesting itself by its action on the magnet, and possibly by auroral pulsations also, of which latter phænomenon it seems excessively difficult to give any other account.-(H.)

ferences from Munich, an arrangement of which it is not very easy to perceive the advantage. The volume in question contains also an investigation by Dr. Lamont of the law of distribution of magnetism in magnetised bars, in which various methods of determining by observation the coefficients of an empirical series representing the intensity of free magnetism in ascending powers of the distance of a point from the centre of the magnet are proposed.

By a communication from M. Boguslawski it appears that in spite of great difficulties arising from want of regular assistants the observations at Prague have been regularly continued, not only on all the term days, but, since January 1st of the current year, also daily at four hours in each day with all the three instruments. Perceptible magnetic disturbances have been noticed by him on January 1, October 6, February 24, March 29 (very great), April 5, May 15, and July 24.

3. Magnetic Surveys.

At the request of the East India Company the magnetic observatories of Simla and Singapore have been supplied with a portable magnetic apparatus, which we hope will be speedily and extensively employed in magnetic surveys having the respective observatories as central points.

M. Kreil is about to add to his most useful observatory labours a magnetic survey of Bohemia, for which he has obtained portable apparatus on the construction proposed by Dr. Lamont.

4. North American Survey.

Letters have been received from Lieut. Lefroy dated from Lachine on the 28th of April, and from Sault Ste Marie, May 20th of the current year, giving an interesting account of his progress so far on his arduous expedition, and detailing his plan of operations, for this and the next year with a sample of each day's performance. Lieut. Lefroy reached Montreal on the 22nd of April, where also his instruments arrived on the 25th (not altogether without injury to the force of his magnets from the extreme badness of the roads). Here, on consultation with Sir G. Simpson, he found it advisable to recast the plan of his route and to resolve on proceeding first to York Fort, and returning thence to Norway House, ascend the Sascatchewan to Edmonton, which he expects to reach on the 20th of September, whence, crossing the Uniga to descend it on the ice to the Slave Lake and return to Athabasca for the remainder of the winter, working his way back to Canada in the next season and taking Moose Fort on the way back. By the adoption of this route a more complete circuit of the focus of maximum intensity will be accomplished than by that originally contemplated. Every necessary order and instruction and every facility he states to have been most readily accorded, and in particular a circular to have been issued to all the officers of the Hudson's Bay Company, amounting to a carte blanche, commanding all the resources of the Company. The line of no variation Lieut. Lefroy states to have been crossed between La Cloche and Sault Ste Marie, up to which point little change of dip had been experienced, his course leading him nearly along the isoclinal line of 77°.

5. Naval Observatories.

The second series of Sir Edward Belcher's magnetic determinations at thirty-two stations, principally at ports in the Pacific Ocean and in the Indian and Chinese Seas, have been reduced by Lieut.-Colonel Sabine and printed in the 2nd Part of the Phil. Trans. for the present year. The two series of Sir Edward Belcher's observations, which are now printed in the Phil. Trans., contain determinations of the three magnetic elements at sixty-one stations

widely distributed over the surface of the globe; and that indefatigable officer has again sailed for the coast of China and the Pacific, furnished with an improved magnetic equipment, including the portable magnetometer apparatus and a Fox's inclinometer and intensity instrument for observations at

sea.

The observations from Captain Blackwood's expedition have begun to arrive, both those with Mr. Fox's instrument at sea and with the portable apparatus on occasions on shore. The observations of this expedition promise to be of great value, from the zeal and intelligence which those already received evince on the part of Lieut. Shadwell and Mr. Evans, under whose particular direction this branch of the public service has been placed by Captain Blackwood.

No expense whatever has been incurred in the present year, but your Committee pray the continuance of the grant made to them at the last meeting to meet such demands as may arise.

Signed on the part of the Committee, J. F. W. HERSCHEL.

Report of the Committee appointed for the Reduction of Meteorological Observations. By Sir J. F. W. HERSCHEL, Bart.

EVERY exertion having been used to complete the series of equinoxial and solstitial observations during the years 1835, 1836, 1837, 1838, whether by writing to parties who have communicated observations, for duplicates of missing series, or by searching the records of observatories, meteorological registers, scientific journals and periodicals, these endeavours have proved so far successful, that at length 334 sets of observations have been collected, made at sixty-nine distinct stations. A synoptic statement of these, with the geographical elements of the stations, the names of the observers or communicators of the observations, and other particulars as far as they could be collected, or are necessary for our present purpose, is annexed to this report. See Appendix (A.).

It will be at once seen on inspection of this synopsis, that although at a few stations (as London, Greenwich, Brussels, Port Louis, Markree, Cadiz) consecutive series, extending over a period of three complete years, have been procured (within the limit assigned to these reductions), yet that this is not the rule but the exception; and that, taken altogether, the observations form anything rather than a connected whole. Under these circumstances, the only point of view which seemed to promise any distinct and definite results, bearing reference to causes prevailing over extensive regions, was that of the barometric fluctuations, considered with a view to the propagation of atmospheric waves, which, it is manifest, can only be traced over any considerable tract of country by this method of inquiry. It is accordingly to these, and to these only, that my attention has been directed; using the observed temperature (where the original observations have not been corrected by the observers) merely as elements of reduction, and referring to the registered state of wind and weather, whenever such reference has been considered elucidatory of any point suggested by the main branch of the inquiry.

If we consider the trifling depth of the ponderable atmosphere regarded as an envelope of the whole globe, the interruption and obstacles offered to its oscillatory movements as a whole by the configuration of the continents and the distribution of mountain chains, but above all, by the vast and capricious variety of local causes affecting the temperatures of particular districts, and

thereby causing partial ascensional and descensional movements and local generations and precipitations of vapour, we shall clearly perceive that, so far as our present purpose is concerned, the particular dimensions and form of our planet have little to do with our inquiry, and that for the immediate purposes of that inquiry we may regard our globe as a plane surface of infinite extent, over particular districts of which systems of oscillation of local origin, and independent of each other, are in progress, and in which we may regard ourselves fortunate if we can now and then succeed in obtaining distinct evidence of the direction, extent, height and velocity of a single wave. The distribution of our stations into groups, grounded on this view of the subject, and the mode of referring the observations of each group to a central station within it, have been described in my report for 1840, and need not therefore here be recapitulated. These groups, it is true, abstractedly considered, are far from those which would be chosen à priori. For example, Mauritius and Van Diemen's Land are but ill adapted to form a group with Indian stations. But for this there is no remedy, and the Mauritius observations (of which, owing to the diligence and zeal of Captain Lloyd, Surveyor-General of that island, we possess a nearly complete series) merit and will receive a separate discussion.

Two objects have been chiefly kept in view in the present inquiry. First, the tracing, where it can be accomplished, the course of one particular wave over the whole area embraced within one of our groups; and secondly, where this cannot be done, the observation of connexions between particular localities with a view to the subdivision of the total area into barometric districts, in which the atmospheric fluctuations shall be, generally speaking, similar in their phases. With these objects the projection of the barometric curves, for all the stations of a group, one sheet for each separate term, has been executed with great care and delicacy by Mr. Birt, and on a scale so large as to allow of the minutest corresponding changes, if any, to be distinctly followed out. The number of sheets so projected is fifty-three, on a scale of an inch to the hour in time, and one inch to 0·066 of barometric altitude; and I must not lose this opportunity of acknowledging many valuable remarks received from that gentleman on the subject of particular cases of much interest, which will be given in his own words, under their several heads as they occur.

I proceed now, therefore, to the discussion of the observations of each term, so reduced and projected, seriatim.

As the whole of the curves of the American group, and seventeen others of the European series, together with their tabulated reductions, have been on a former occasion submitted to the inspection of the meeting, to the number of 105, it has not been thought necessary to risk the loss or damage of the remainder by transmitting them herewith, they being in all respects

similar.

European Group.

June 1835. This term affords only two series, those of London and Brussels. The projected curves exhibit a pretty regular descent, and tolerably well-maintained parallelism during the first sixteen projected hours, or until 3 A.M. of the 22nd, when the Brussels curve attains a minimum, and pretty abruptly ascends again to the end of the series. The London curve, on the other hand, continues to descend till 11 A.M. of the 22nd, where it also attains a minimum, and begins to reascend. If this minimum represent, as it probably does, the trough of a barometric_wave which at 3 A.M. was vertically over Brussels, and at 11 A.M. over London, the wave must have been