distinct traces were left of their progrefs through the air. During the explosion of a meteor, on the 20th August 1789, near Bordeaux, a ftone, about fifteen inches diameter, broke through the roof a cottage, and killed a herdsman and fome cattle. Part of the ftone is now in the museum of Mr Greville, and the rest in that of Bordeaux. It is fingular that this fact is not mentioned by M. Izarn, nor by Vauquelin, although he examined a fpecimen evidently taken from the fame ftone, and received a procès-verbal of the manner in which it fell. We take the account from Mr Greville's paper, (Phil. Tranf. 1803. part I.); and he appears to have received it from M. St Amand, Profeffor of Natural History at the Central School of Agen.

It is quite impoffible, we apprehend, to deny very great weight to all thefe teftimonies; fome of them given by intelligent eyewitneffes; others by people of lefs information, indeed, but prepoffeffed with no theory; all concurring in their defcriptions; and examined by various perfons of acuteness and respectability, immediately after the phenomena had been exhibited. Without offering any farther remarks, then, upon this mafs of external evidence, we shall only remind our readers of the main points which it seems fatisfactorily to fubftantiate. It proves, that, in various parts of the world, luminous meteors have been seen moving through the air, in a direction more or lefs oblique, accompanied by a noife, generally like the hifling of large fhot, followed by explofion, and the fall of hard, ftony, or femimetallic maffes, in a heated state. The hiffing found, fo univerfally mentioned; the fact of ftones being found, unlike all those in the neighbourhood, at the fpots towards which the luminous body or its fragments were feen to move; the fcattering or ploughing up of the foil at those fpots, always in proportion to the fize of the ftones; the concuffion of the neighbouring ground at the time; and, above all, the impinging of the ftones upon bodies fomewhat removed from the earth, or lying loofe upon its furface-are circumftances perfectly well authenticated in thefe reports; and, when taken together, are obviously fatal to any theory, either of the maffes having previoufly exifted in the foil ready formed, and having been difclofed by the electric. fluid-or of their component parts having existed there, and having been united and confolidated by that fluid.

II. While the internal evidence on this queftion, that is, the inference arifing from an examination of the ftones themselves, agrees moft harmonioufly with the conclufion to which the narratives above analyzed force our affent, and greatly strengthens that conclufion, it alfo leads to a farther knowledge of the fub

jec,

ject, than the mere external evidence could of itself have afforded

us.

The reports from all thofe who obferved the meteors, and found the ftones in the neighbourhood, after the explosions, agree in defcribing thofe fubitances as different from all the furrounding bodies, and as prefenting, in every cafe, the fame external appearance of femi-metallic matter, coated on the outfide with a thin black cruft, and bearing strong marks of recent fufion. This general resemblance we fhould be perfectly entitled to infer from the various accounts of eye-witneffes, even if no more particular obfervations had been made by men of fcience, to whofe inspection many of the fallen bodies were fubmitted. But fortunately a confiderable number of these fingular fubftances have been examined, with the greatest care, by the first chemifts and naturalifts of the age; and their inveftigations have put us in poffeffion of a mass of information, capable of convincing the moft fcrupulous inquirer, that the bodies in question have a common origin, and that we are as yet wholly unacquainted with any natural procefs which could have formed them on our globe.

M. De la Lande appears to have examined the ftones which fell near Bourg, in the province of Breffe, 1753, with fome attention. He remarks their external coating of black vitrified matter, the metallic or pyritical threads interfperfed through them, and more particularly the cracks filled with metallic particles. His chemical analyfis is very meagre and unfatisfactory; but fuch as it was, its refults, as well as the general obfervations of external character, corresponded with the inferences drawn by him from a fimilar examination of the ftone which fell, in 1750, near Coutances, in Normandy, at the distance of three hundred and fixty miles from Bourg.

The external appearance of the three ftones prefented to the Academy of Sciences, as having fallen in different parts of France during the year 1768, was precifely the fame. But Meffrs Lavoifier, &c. the committee appointed to examine them, performed the chemical analysis with much greater accuracy and fulness than M. De la Lande had done. That which fell in the Maine, and was prefented by the Abbe Bachelay, underwent the most careful procefs. It was found to contain, of fulphur, 8 per cent.; iron, 36; and vitrifiable earth, 55. It must be remarked, however, that this decompofition was effected by means of experiments performed upon an integral part of the whole ftone, confidered as a homogeneous fubftance; whereas, it is in fact a congeries of fubftances, which ought to have been feparately analyzed. This confideration will, ia part at least,

enable

enable us to account for the apparent difcrepancy between the refults obtained by the academicians and thofe of later experimentalifts. Meffrs Lavoifier, &c. alfo examined particularly another stone, faid to have fallen in a different part of France, and obtained very nearly the fame refults. The only difference was, that it did not give out fulphurated hydrogenous gas when acted upon by the muriatic acid; a peculiarity diftinctly obferv able in the other fubftance.

The defcription which Profeffor Barthold gives of the external character of the ftone which fell near Enfifheim, in the fifteenth century, correfponds exactly with the defcriptions given of these ftones, and of the ores examined by M. De la Lande. The refults of his analyfis are somewhat different; but he examined the whole heterogeneous compound, and not the parts feparately. He concluded, that this mass contained 2 per cent. of fulphur, 20 of iron, 14 magnesia, 17 alumina, 2 lime, 42 filica. Mr Howard has very juftly remarked, that the Profeffor's own account of his experiments is at variance with the idea of lime being contained in the fubftance; and that he has given no fufficient proof of the existence of alumina. It is alfo to be obferved, that from the exceptionable method of analysis pursued both by Barthold and the academicians, the metallic particles were not examined with fufficient precifion. The specific gravity of the ftones examined by the academicians was to that of water, as 3535 to 1000. The fpecific gravity of the ftone of Enfifheim, as tried by Barthold, was 3233; that of the stone examined by Gaffendi (who faw it fall) was 14, common marble being 1; and, taking the specific gravity of marble to that of water, as 2716 to 1000, the fpecific gravity of the ftone obferved by Gaffendi will be to that of water as 3456 to 1000. So near a coincidence between obfervations, made at fuch a distance of time, upon these various fubftances, cannot fail to ftrike us as very remarkable, and to prepare us for that fuller demonstration of their identity, which was referved for the labours of our countryman Mr Howard.

This excellent philofopher has elucidated the fubject of our prefent confideration, by a courfe of experiments as interefting and instructive as any that the fcience of chemical analyfis can boat of. He fortunately obtained fpecimens of the ftones which fell in several very diftant quarters of the globe; at Benares, and in Yorkshire (as we have already defcribed); near Sienna, and in Bohemia, according to evidence not altogether fo fatisfactory as that upon which the other narratives reft.

He began his inquiries, very judicioufly, by a minute examination of the external mineralogical characters of these four

fubftances;

fubftances; and in this part of his task he was indebted to the learning and expertnefs of the Count de Bournon. The fubftances were found to resemble each other very closely in their general appearances, and in the nature of their component parts. The chief difference confifted in the different proportions in which the fame component parts were combined, fo as to form the aggregate of the heterogeneous maffes. Their specific gravities were nearly the fame, unless that the abundance of iron in one of the maffes caused a confiderable increase of its gravity. It may contribute to the formation of a precife eftimate, if we prefent, in one view, the refults of the experiments made to measure the specific gravities of the moft remarkable specimens hitherto examined. The four laft in the lift were calculated by the Count de Bournon. The specific gravity of water being 1000, that of the Enfifheim stone is

Gaffendi's*

Bachelay'st

Yorkshire

Sienna

3233

3456

3535

3508

3418

3352

4281

Benares

Bohemia

All the stones examined by Count de Bournon and Mr Howard were found to consist of four distinct substances; fmall metallic particles; a peculiar martial pyrites; a number of globular and elliptical bodies, alfo of a peculiar nature; and an earthy cement furrounding the other conftituent parts. It was only the ftone from Benares that Mr Howard could separate into its constituent parts, with fufficient accuracy, and in fufficient abundance, for a minute analysis of each. He found, however, that the nature of the metallic particles was the fame in all; they were in each cafe an alloy of iron and nickel. In the pyrites of the Benares ftone, nickel as well as iron was detected; and the eafy decompofition of the pyrites by muriatic acid, in all the fpecimens, afforded a diftinguishing character of this fubftance. The globules in the Benares ftone contained filica, magnefia, and oxides of nickel and iron; the earthy cement confifted of the fame fubftances, very nearly in the fame proportions. In the other ftones, thefe globules could not be easily feparated from the cement and pyrites. Mr Howard, therefore, after freeing the aggregate as well as poffible from the metallic particles, and feveral of the globules, was obliged to fatisfy himself with analyzing the heterogeneous mafs. Still the compofition appeared wonderfully to agree with that of the bafis and globules of the

Found in Provence.

+ Found in the Maine.

the Benares stone; as the following Table, collected from Mr Howard's experiments, and reduced to the parts of a hundred, will clearly evince.

About the time that Mr Howard was engaged in these interesting researches, and before he had publifhed the refult of them, M. Vauquelin happened alfo to be occupied with the very fame subject. He analyzed, though by a different procefs, the Benares ftone, and two others which fell in 1789 and 1790 in the fouth of France. The refults of his experiments agreed with thofe of our diftinguished countryman in every particular; and we are now entitled to conclude, with perfect confidence, that the ftones which have at different times fallen upon the earth, in England, France, Italy, and the East Indies, are precifely of the fame nature, confifting of the fame fimple fubftances arranged in fimilar compounds, nearly in the fame proportions, and combined in the fame manner, fo as to form heterogeneous aggregates whose general refemblance to each other is complete. We are further warranted in another important inference, that no other bodies have as yet been difcovered on our globe, which contain the fame ingredients; and, more particularly, that the analysis of these stones has made us acquainted with a fpecies of pyrites not formerly known, nor any where elfe to be found.

The general analogy between these stones and the maffes of native iron found in different parts of the world, was too ftriking to efcape the eminent inquirers who have investigated this fubject. They refemble each other in their external character, though not by any means fo closely as the ftones; but in one circumftance of their chemical compofition, they have a remarkable similarity, both among themselves, and towards the ftony fubftances. M. Prouft, a confiderable time before the date of Mr Howard's difcoveries, had proved that the enormous mafs of native iron found in South America, contained a large portion of nickel in its compofition. Mr Howard was led to the fame conclufion by analyzing another portion of this body; and he found that the folitary maffes difcovered in Siberia, Bohemia, and Senegal, contained a mixture of the fame metal with iron, though in various proportions. The Bo