with what is known in the case of solids, that in gaseous media also there is equality in the powers of radiation and absorption. Bodies which exert an absorbent effect in the liquid form preserve it in the gaseous state. If further experiments should confirm Prof. Tyndall's views upon the absorptive action of aqueous vapour upon radiant heat of low intensity, these results must materially modify some of the views hitherto held upon the meteorological relations of aqueous vapour.

The Bakerian Lecture, by Mr. Sorby, is entitled by him "On the Direct Correlation of Mechanical and Chemical Forces." In this paper are embodied a series of observations upon the influence of pressure upon the solubility of salts, in which he has obtained results analogous to the changes observed in the freezing-point of liquids under pressure. He finds, in cases where, as is usual, the volume of the water and the salt is less than the volume of the water and the salt separately, that the solubility is increased by pressure, but that, in cases where (as when sal-ammoniac is dissolved in water) the bulk of the solution is greater than that of the water and salt taken separately, the solubility is lessened by a small but measurable amount. On the contrary, salts which expand in crystallizing from solution must, under pressure, overcome mechanical resistance in that change; and as this resistance is opposed to the force of crystallization, the salt is rendered more soluble. The extent of the influence of pressure, and the mechanical value of the force of crystalline polarity, were found to vary in different salts. Mr. Sorby also indicates the results of the action of salts upon certain carbonates under pressure, and purposes pursuing his researches upon chemical action under pressure. This paper may therefore be regarded as the first of a series upon a highly interesting and important branch of investigation, for which Mr. Sorby appears to be specially fitted, from his combining the needful geological knowledge with the skill in manipulation required in the physical and chemical part of the inquiry.

The examination of the bright lines in the spectra of electric discharges passing through various gases, and between electrodes of various metals, has of late years attracted very general attention. Each elementary gas and each metal shows certain well-marked characteristic lines, from the presence or absence of which it is commonly assumed that the presence or absence of the element in question may be inferred. But the question may fairly be asked, Has it been established that these lines depend so absolutely on chemical character that none of them can be common to two or more different bodies? Has it been ascertained that, while the chemical nature of the bodies remains unchanged, the lines never vary if the circumstances of mass, density, &c. are changed? What evidence have we that spectra are superposed, so that we observe the full sum of the spectra which the electrodes and the medium would produce separately?

To examine these and similar questions in the only unimpeachable way

(that of actual experiment) formed the object of a long and laborious research by Dr. Robinson, the results of which are contained in a paper in our Transactions. In the course of this research, Dr. Robinson had occasion to take careful measures of the positions of all the bright lines visible (and not too weak to measure) in a great number of spectra-those, namely, of the induction discharge passing between electrodes of twenty different metals, as well as graphite, most of which were observed in each of five different gases (including air), and for each gas separately at the atmospheric pressure and at the low pressure obtained by a good air-pump.

On taking an impartial survey of this great assemblage of experimental facts, Dr. Robinson inclines to the opinion that the origin of the lines is to be referred to some yet undiscovered relation between matter in general and the transfer of electric action; and that while the places of the lines are thus determined independently of particular circumstances, the brightness of the lines is modified, according to the special properties of the molecules which are present, through a range from great intensity down to a faintness which may elude our most powerful means of observation.

By a discussion of the results of the magnetic observations maintained for several years past at the Kew Observatory with an accuracy previously unattained, and by combining these with the earlier results of the observations at the British colonial observatories, I have been enabled to trace and, as I believe, satisfactorily to establish the existence of an annual variation in the three elements of the earth's magnetism, which has every appearance of being dependent upon the earth's position in her orbit relatively to the sun. Substantiated by the concurrent testimony of observations in both hemispheres, and in parts of the globe most widely distant from each other, this conclusion furnishes an additional evidence of a cosmical magnetic relation subsisting between the earth and other bodies of the solar system, and thus extends the scope and widens the basis of sound induction upon which the permanent relations of magnetical science must rest.

To Dr. Otto Torell, Professor of Zoology in the University of Lund, we are indebted for a communication of much interest, informing us of the progress made by an expedition appointed by the Swedish Government at the recommendation of the Royal Academy of Sciences at Stockholm, to execute a survey preliminary to the measurement of an arc of the meridian at Spitzbergen. The objects of the preliminary survey were to ascertain whether suitable angular points for a triangulation could be found from Ross Island at the extreme north, to Hope Island at the extreme south of Spitzbergen, and to determine on a favourable locality for the measurement of a base-line. The result of the first year's exploration has been the selection of stations, on hills of moderate height and easy access from the coast, for nine triangles shown in the sketch accompanying Dr. Torell's paper, including Ross Island in the extreme north, and extending

over about 1° 50′ of the proposed arc of 4 degrees. A convenient locality has also been found for the base-line. The continuation of the preliminary survey to the extreme southern limit is to be the work of the summer of 1864. The report of the Geodesical Surveyors has shown that the northern portion presents no impediments which may not be surmounted by courage and perseverance; and with regard to the southern portion, the knowledge already acquired is considered to justify the expectation that the result of the second year's exploration will be no less favourable. Should such be the case, it is anticipated that the necessary steps will be taken for carrying into execution the measurement of the arc itself.

I may perhaps be permitted to allude for a moment to the peculiar interest with which I must naturally regard the proposed undertaking. The measurement of an arc of the meridian at Spitzbergen is an enterprise which nearly forty years ago was a cherished project of my own, which I had planned the means of executing, and which I ardently desired to be permitted to carry out personally. I may well therefore feel a peculiar pleasure in now seeing it renewed under what I regard as yet more promising auspices,-whilst I cannot but be sensible of how little I could have anticipated that I should have had the opportunity, at this distance of time, and from this honourable chair, of congratulating the Swedish Government and Academy upon their undertaking, and of thanking Dr. Torell for having traced its origination to my early proposition.

It is well remarked by Dr. Torell, that the triangulation, should it be proceeded with, will not be the only result of the years of scientific labour at Spitzbergen. There are, indeed, many important investigations for which the geographical circumstances would be eminently favourable. Two such may be specified, for which we may reasonably anticipate that full opportunity would be afforded, and for which the requisite instruments of precision are neither costly nor cumbersome. One is a more exact determination of the data on which our Tables of Astronomical Refraction are founded. The other is the employment of Cagnoli's method for determining the figure of the earth by occultations of the fixed stars*. This last would be tried under circumstances far more favourable than those contemplated by its original proposer, by reason of the high latitude of the northern observer-the greater number of stars in the moon's path, now included in our catalogues, of which a special ephemeris might be made -and the much greater amount of concerted corresponding observations which might now be secured. The advantage peculiar to this mode of determination is, that it is exempt from the influence of local irregularities in the direction and force of gravity which embarrass the results of

* Antonio Cagnoli, “Nuovo e sicuro mezzo per riconoscere la Figura della Terra,” Memorie della Società Italiana, Verona, vol. vi. 1792.

An English translation, with Notes and an Appendix, was printed for private circulation in 1819, by Mr. Francis Baily.

the measurements of degrees and of pendulum experiments. As a third and thoroughly distinct method of investigation, it seems at least well deserving of a trial.

Swedish naturalists are not likely to undervalue the interest attaching to careful examinations of the constancy or variation of the elevation of land above the sea-level; and I may therefore venture to refer them to a paper in the Phil. Trans. for 1824 (Art. xvi.), written from Spitzbergen itself in July 1823, containing the particulars of a barometrical and trigonometrical determination of the height (approximately 1644 English feet) of the well-defined summit of a conspicuous hill in the vicinity of Fairhaven. The barometrical comparison was repeated on several days, the barometer on the summit of the hill being stationary, and the observation of the two barometers strictly simultaneous, the stations being visible from each other by a telescope. The height as given by the two methods, barometrical and trigonometrical, was in excellent accord. The hill may be identified with certainty by the plan which accompanies the paper referred to: it is of easy access, and may be remeasured with little difficulty.

It will be remembered that a few years ago the attention of the Royal Society was called by the Foreign Office to the circumstance of several glass bottles with closed necks having been found on the shores of the west coast of Nova Zembla, leading to a conjecture that they might afford some clue to the discovery of the missing ships of Sir John Franklin's Expedition. The inquiries instituted by the Royal Society traced the bottles in question to a recent manufacture in Norway, where they are used as floats to the fishing-nets employed on that coast. These floats, accidentally separated from the nets, had been carried by the streamcurrent which sets along the Norwegian coast round the North Cape, and thus afforded evidence of the prolongation of the current to Nova Zembla. The Swedish Expedition, in the course of its summer exploration, found on the northern shore of Spitzbergen several more of these bottle-floats, some of which even bore Norwegian marks and names, supplying evidence, of considerable geographical interest, of the extension of the Norwegian stream-current to Spitzbergen, either by a circuitous course past the shores of Nova Zembla, or by a more direct offshoot of which no previous knowledge existed. It is thus that step by step we improve our knowledge of the currents which convey the waters of the more temperate regions to the Polar seas and produce effects which are traceable in many departments of physical geography.

The application of gun-cotton to warlike purposes and engineering operations, and the recent improvements in its manufacture, have been the ubject of a Report prepared by a joint Committee of the Chemical and Mechanical Sections of the British Association, consisting chiefly of Fellows of the Royal Society. The Report was presented at the Meeting in New

castle in September last, and is now in the press. The Committee had the advantage of personal communication with General von Lenk, of the Imperial Austrian Artillery, the inventor of the system of preparation and adaptation by which gun-cotton has been made practically available for warlike purposes in the Austrian service. On the invitation of the Committee, and with the very liberal permission of the Emperor of Austria, General von Lenk visited England for the purpose of thoroughly explaining his system; and we have in the Report of the Committee the information, thus gained directly from the fountain-head, of the results of his experience in the course of trials extending over many years, together with additional investigations by individual members of the Committee.

The advantages which are claimed for gun-cotton over gunpowder for ordnance-purposes and mining-operations are so many and so important as to call imperatively for the fullest investigation. Such an inquiry, however, in its complete sense, is both beyond and beside the scope and purposes of a purely scientific body; and the British Association have done well (whilst reappointing the Committee to complete certain experiments which they had devised with the view of clearing up some scientific points which are still more or less obscure) in pressing on the attention of Her Majesty's Government the expediency of instituting under its own auspices a full and searching inquiry into the possible applications of gun-cotton in the public service.

The absence of smoke, and the entire freedom from the fouling of the gun, are points of great moment in promoting the rapidity of fire and the accuracy of aim of guns employed in casemates or in the between decks of ships of war; to these we must add the innocuous character of the products of combustion in comparison with those of gunpowder, and the far inferior heat imparted to the gun itself by repeated and rapid discharges. With equal projectile effects, the weight of the charge of gun-cotton is but one-third of that of gunpowder; the recoil is stated to be reduced in the proportion of 2 to 3, and the length of the gun itself to admit of a diminution of nearly one-third. These conclusions are based on the evidence of long and apparently very carefully conducted courses of experiment in the Imperial Factory in the neighbourhood of Vienna. The results appear to be especially deserving the attention of those who are engaged in the important problems of facilitating the employment of guns of large calibre and of great projectile force in the broadsides of our line-of-battle ships, and in reducing, as far as may be possible, the dimensions of the ports.

In the varied applications of explosive force in military or civil engineering, the details of many experiments which bear on this branch of the inquiry are stated in the Report of the Committee, and appear to he highly worthy of consideration and of further experiment.

It cannot be said that the advantages now claimed for gun-cotton are altogether a novel subject of discussion in this country. When the