none could now be procured that diminished so much as the parcel originally made ufe of. After various researches and experiments, dictated by a thorough knowledge of the subject, and which have difcovered many curious particulars refpecting the properties of this class of earths, he has recourse to an admixture of the earth of alum, the ingredient on which the diminution by fire and all the argillaceous properties depend; and by a due proportion of which the common porcelain clays of Cornwall are made to correfpond in all degrees of heat with the original clay, receiving from it at the fame time fome other important advantages. Coincidence with the original,' he obferves, was not indeed effential; but as many degrecs of heat were already before the public, measured by thermometer-pieces made of the firft clay, and as the correspondence of the firft with Fahrenheit's fcale, had likewife been in fome measure afcertained *, it was defirable that the fame degrees of heat fhould continue to be expreffed by the fame numbers." Of the embarraffing properties of fome of the natural clays, we fhall mention one, on account of what appears to us a very important confequence refulting from it. Though they con tinued diminishing with tolerable regularity, up to a certain period of heat, about that in which caft iron melts, yet many of the pieces, urged with a heat known to be greater than that, were found not to be diminished fo much as those which had fuffered only that lower heat. Further experiments fhewed, that after diminishing to a certain point, they begin, upon an increase of the heat beyond that point, to fwell again and as, this effect is conftant in certain clays, and begins earliest in those which are most vitrifcible, and as clays are found to fwell upon the approach of vitrification, I look upon this enlargement of bulk, however inconfiderable, as a fure indication of the clay having gone beyond the true porcelain ftate, and of a difpofition taking place towards vitrification.-The degree of heat, therefore, at which this enlargement begins, may be confidered as a criterion of the degree of vitrifcibility of the compofition; which points out a new ufe of this thermometer, enabling us to afcertain the degree of vitrifcibility of bodies that cannot actually be vitrified by any fires which our furnaces are capable of producing.' Obfervations on the Affinities of Subftances in Spirit of Wine. By John Elliot, M. D. This Paper, in the form of a letter to Mr. Kirwan, contains fome experiments in confirmation of a pofition which the Author had advanced in a former publication, that certain decompofitions will take place in fpirit of wine, which will not at See Rev. vol. lxxii. p. 250. all in water, nor in the dry way.' The particular decompofition here treated of, is that of diachylum plafter (a compound of litharge and oil) with fea-falt, by boiling them in fpirit of wine: the acid of the fea falt unites with the litharge, and its alcali with the oil, the two latter forming together a true foap, which diffolves in the fpirit, and may be obtained in its proper form by evaporation. As no feparation takes place in water, it follows, that the apparent affinities depend, not folely upon the attractions of the ingredients to one another, but in part upon their attractions to the liquid employed as an intermedium. In fpirit of turpentine, the affinities were ftill further diverfified; the diachylum diffolved, and the common falt remained at the bottom. New Experiments upon Heat. By Colonel Sir Benjamin Thompfon, Knt. F. R. S. From the ftriking analogy between the electric fluid and heat, in respect to their conductors and non-conductors (bodies which are good conductors of the one being generally fo of the other alfo), it was natural to imagine that the Torricellian vacuum, which affords fo ready a paffage to the electric fluid *, would do the fame to heat. But a feries of curious, and admirably wellcontrived experiments, detailed in this Paper, prove, without leaving a fhadow of doubt, that the vacuum conducts heat far more fluggishly than air;-that, neverthelefs, air of different denfities differs little in this refpect;-but that the conducting power of air is very greatly increafed by humidity. On this remarkable effect of moisture in the air, the Author remarks, with what infinite wisdom and goodness Providence appears to have guarded us against the ill-confequences of exceffive heat and cold in the atmosphere: for were it poffible for the air to be as damp during the fevere colds of the winter months, as it fometimes is in fummer, its power of conducting the heat from our bodies, and of courfe its apparent coldness to us, would be intolerable; but happily its power of holding water in folution, and therewith its power of robbing us of our animal heat, diminishes in proportion as its coldness is increased. If colds or catarrhs are occafioned by our bodies being robbed of heat, the reason is plain why thofe diforders prevail moft during the cold autumnal rains, and on the breaking up of the froft in fpring. It is hence alfo plain, why damp houfes, and damp beds, are Though it has lately been difcovered, that the perfect vacuum does not conduct electricity, this does not affect the Author's comparifons; for the vacuum, in his experiments, was fuch as may be prefued to have been a conductor; and in either cafe, the difference, in that refpect, between the electric fluid and heat is equally Striking. fo dangerous; and why the evening air is fo pernicious in fummer and autumn, and not fo in the hard frofts of winter. Phyficians have been puzzled to account for the generation of the extraordinary quantity of heat fuppofed to be carried off from animal bodies by the cold air in winter, above what they communicate to the warmer atmosphere in fummer; but it is more than probable, that the difference in the quantities of heat fo loft or communicated, is infinitely lefs than they have imagined. The Author examines alfo the conducting power of mercury, and finds it to be greater than that of water, in the proportion of about 100 to 31. Hence it is plain why mercury appears fo much hotter or colder to the touch than water does, though really of the fame temperature; for the intenfity of thofe fenfations does not depend entirely upon the degree of heat in the body exciting them, but on the quantity of heat which it can communicate to or from us in a given short period, or the intensity of the communication. Sir Benjamin propofes continuing his experiments on heat; and experiments fo accurately and judiciously conducted, on fo interesting and obfcure a subject, cannot fail to afford important results. An Account of Experiments made by Mr. John McNab, at HenleyHoufe, Hudson's Bay, relating to freezing Mixtures. By Henry Cavendish, Efq; F. R. S. and A. S. Thefe experiments were made at Mr. Cavendish's defire, with materials provided and adjusted by him, in order to afcertain fome particulars in his remarks fubjoined to Mr. Hutchins's Paper in the 73d volume of the Tranfactions. As heat is gene-. rated in the congelation of fluids, and cold in the liquefaction of folids; and as the cold produced by mixing fnow with fpirit of nitre is fupposed to be owing merely to the liquefaction of the fnow, it should follow, that there may be a degree of cold, in which the nitrous fpirit, fo far from diffolving fnow, will fuffer part of its own water to freeze; and in that cafe no additional cold fhould be produced by the mixture of fnow with it. As ftrong fpirit of nitre generates heat with water, it does the fame with the water that is formed on the fift addition of fnow to it, and no cold is produced till the fnow amounts to about one fourth of the weight of the acid; it fhould follow, that if the acid be diluted at first with one fourth of water, no heat will be ge * Mr. Cavendish adopts Sir Ifaac Newton's idea, that heat is not a distinct body, but a quality, produced by the inteftine action of the parts of bodies. So far as relates to the explication of thefe experiments, the theory affects only the mode of expreffion; inftead of say-ing that heat is generated, it is now more cuflomary to fay that it is dijengaged, or let loose. nerated, nerated, and the cold will of course be greater. The experiments are perfectly fatisfactory, and afcertain fome other interefting particulars, in a region of fcience which philofophers have but few opportunities of exploring. It appears from thefe experiments, that the nitrous acid is not only fufceptible of an aqueous congelation, or freezing of the watery part, but of a fpirituous, or freezing of the acid itself:That when cooled to the point of aqueous congelation, it has no tendency to diffolve fnow, and thereby produce cold, but on the contrary, is difpofed to part with its own water :-That the tendency to diffolve fnow, and produce cold, is by no means deftroyed by its being cooled to the point of fpirituous congelation, or even actually congealed:-That both the strong and diluted acids bear, like water, to be cooled very much below their freezing point before the congelation begins, and rife up to their freezing points as foon as it does begin:-That, contrary to water, they fhrink in freezing, very much, the furface becoming depreffed, and full of cracks, and the ice finking freely in the unfrozen fluid :-And that their freezing point varies according to a very unexpected law, the acid of a certain degree of ftrength freezing much easier than that which is either ftronger or weaker. The loweft heat in which any nitrous fpirit was found to freeze, was 1 below o of Fahrenheit: the ftrength of this fpirit was 411, that is, 1000 parts of it would diffolve 411 of marble; and when any ftronger or weaker fpirit was frozen in part, the frozen part approached nearer to that ftrength than the unfrozen. The general refults of the experiments on this head are expreffed in the annexed table. The first column fhews the ftrengths of the acids; and the fecond the degree below o at which they respectively froze. 54031 411 1 -380 45 } 243447 210 17 fpirituous congelation. } aqueous congelation, The vitriolic acid contracts in freezing, as the nitrous does, its ice finking to the bottom of the unfrozen fluid; but it does not appear to have any point of easiest congelation, feeming rather to freeze eafier and eafier, the more concentrated it is, without limitation. It feems alfo, that certain parts are more congealable than the reft, and that the difference does not depend altogether on ftrength, but on fome lefs obvious quality, probably on that, whatever it is, which forms the difference between glacial and common cil of vitrioļ. Abstract of a Register of the Barometer, Thermometer, and Rain, at Lyndon in Rutland, in 1785. By Tho. Barker, Elq. &c. The depth of rain at Lyndon was about 20 inches; at South Lambeth it was only 19, at Fyfield 241, and at Selbourn 311. From fome obfervations iubjoined to this regifter it appears, that the annual quantity of rain is very variable in the fame place at different periods. At Lyndon, from 1740 to 1743 inclufive, the mean depth was only 16 inches in a year, and yet no complaint was made of dry fummers in any of those years; the fummers were fhowery, but the winters dry. From 1741 to 1750, the mean depth was 18 inches; from 1751 to 1760 22-1761 to 1770, 23;-1770 to 1780, 26. In three of the years of this laft period, 1773, 4, and 5, the mean annual depth was 32 inches; and in one of them, from October 1773 to September 1774, the depth was 391. It is plain from thefe obfervations, how little dependence can be had upon average quantities taken on periods of fmall extent. Magnetical Experiments and Obfervations. By Mr. Tiberius Cavallo, F. R. S. (The Lecture founded by the late Henry Baker, Efq. F. R. S.) Thefe experiments relate chiefly to the magnetic property which has lately been obferved in fome kinds of brafs. The general refult of them is, That moft brafs becomes magnetic by hammering: That the magnetifm is deftroyed by annealing, or foftening in the fire, reftored again by hammering, and thus alternately, as often as the hammering and annealing have been repeated: That the magnetifm is not owing to any particles of iron communicated by the tools, for it is producible by beating the brafs between pieces of flint or copper; and that the deftruction of the magnetifm is not owing to the calcination of any iron particles, for it takes place though the brafs be furrounded with charcoal powder in a clofe crucible, Mr. Cavallo is therefore of opinion, that the magnetifm acquired by brafs is not owing to any iron in it, but to fome particular configuration of its component parts, occafioned by the ham mering. There are indeed pieces of brafs which have vifible particles of iron in them, but these are magnetic in their foft as well as hard ftate; and there are others, which cannot, by any known means, be rendered magnetic at all. A propofition fo fingular and important, as the existence of magnetism, or the power of attraction to the magnet, independent of iron, is certainly not to be admitted without rigorous examination. The Author himself, in a Poftfcript to the Lecture, ftarts an objection, and gives fome experiments to obviate it. The brafs may owe its magnetifm to an irony matter intermixed, and this iron may be magnetic, or not magnetic, according as it is |