Prof. M. A. Usov.

TOMSK METEORITE.

(Petrographic study).

Tomsk.
Steam Printing House of L. K. Orlova, Yamskaya St., own house.
1916.

Prof. M. A. Usov.

THE TOMSK METEORITE

(A petrographic study).

In the middle days of June 1912, in the vicinity of a village of Tomsk Uezd, at about 5 o’clock in the evening, a meteorite fell. Its fall was accompanied by a cracking sound and light phenomena, and was noticed by peasants mowing nearby. Rushing to the place where the meteorite had fallen, the peasants easily found it by the thin vapor issuing from a little pit in the heavily waterlogged soil. The meteorite, however, was not at the bottom of this pit, and they had to dig down into the boggy ground almost a sazhen in order to get it. When the meteorite was dug out, its surface was still hot, and it was not easy to take it in the hands.

The peasants decided to divide this heavenly gift among themselves and, with blows of an axe, broke the meteorite into 5 pieces. Each participant in such an extraordinary division, taking a piece of the sacred stone, later placed it at home on the icon shelf. Only one of the peasants, who proved to be the elder of that village, considered it not useless to present his piece as a gift to the Tomsk Governor, who at that time was Mr. Shteven. When Mr. Shteven died, the piece of meteorite passed to K. Zenkov, and the latter handed it over to the well-known Tomsk physician Ya. I. Beigel, who finally placed the meteorite at my disposal and communicated to me the above data on its origin. I consider it my pleasant duty to express here my gratitude to Ya. I. Beigel for this gift, which gives me the opportunity to make a description of one of the celestial wanderers that so rarely, in general, fall upon the surface of our planet.

Thus, the piece at my disposal represents only part of a meteorite that fell somewhere not far from the city of Tomsk (in any case in Tomsk Uezd), and for that reason has been provisionally named by me the Tomsk meteorite; it is possible that later it will be possible to establish more precisely the place where this meteorite fell and the details of the phenomenon itself, but for the present it is necessary to confine oneself to the information communicated on this matter—information that is scanty and not entirely reliable.

The body of the meteorite has an irregular—parallelepipedal—shape, with dimensions of approximately \(8 \times 6 \times 5\) cm.; only two surfaces of this body are primary, being covered with a typical fusion crust; on the other sides it is bounded by fracture surfaces produced when the meteorite was broken. The weight of the piece, after fragments had been separated for the preparation of thin sections and for chemical analysis, proved to be 795 g., with a specific gravity of 3.624. In accordance with the latter value and with its composition, the meteorite should be assigned to the stony meteorites according to E. Cohen’s classification *).

In fresh fracture the meteorite described is colored ash-gray, a color that appears denser under artificial illumination and is due, in essence, to the composition of the meteorite from silicate formations of corresponding coloration. In view of such a composition and, in general, of its very fine-grained constitution, the aggregate has a lithoid appearance on a rough fracture, differing in these rather important features from the rocks of the earth’s lithosphere.

Above all, characteristic of the given meteorite is the presence of chondrules. These chondrules are almost spherical bodies consisting of radial-fibrous accumulations of thin prismatic individuals of a light-gray silicate with a greenish-yellow tint. Only in the comparatively very rarely encountered large chondrules, up to 1 cm in diameter, can it be clearly seen that the radial-fibrous accumulations are situated within the body generally eccentrically, having the appearance of sectors, but sometimes they occur in a more or less arbitrary relation to one another. Usually, however, the chondrules have considerably smaller dimensions, and then they are almost indistinguishable from individualized segregations of silicates, which likewise for the most part possess slightly rounded forms and a dull fracture. In view of this, chondrules in general do not stand out sharply and cannot be knocked out of the meteorite, which therefore belongs to the group of intermediate chondrites **).

Another characteristic feature of the meteorite, distinguishing it from our igneous rocks, is the absence of a clearly expressed structure. Only upon careful examination of the specimen can it be established that the meteorite is an aggregate of grains, silicates of several varying shades of light-gray color, but its appearance

) E. Cohen. Meteoritenkunde II.—Stuttgart, 1905: 22.
) E. Cohen.* Meteoritenkunde II.—Stuttgart, 1903: 65.

the grains in general are very indistinct; it is difficult even to determine precisely what dimensions they have, and only here and there do individuals stand out quite definitely, especially in the presence of slightly shining cleavage planes, acquiring the character of segregations up to 2 mm in size. And since these segregations not infrequently have an irregular form and varied coloration and at the same time are with difficulty distinguishable from the chondrules, which are so alien in structure, the whole formation acquires, as it were, a fine-brecciated or tuffaceous texture, which is, indeed, characteristic of intermediate and gray chondrites.

The original texture of the meteorite is emphasized by the very uneven distribution of fine-grained metallic alloys and pyritic formations, which play the role of essential impurities in the aggregate. For the most part these components occur as inclusions in the meteorite, but sometimes they stand out as thin short veinlets and small accumulations. In their mode of occurrence the metallic compounds are analogous to the silicate constituent parts of the meteorite, and, if they are sharply separated from the latter, this is due to their external properties.

It is difficult to determine macroscopically what the origin of the meteorite was in comparison with the rocks of the Earth’s lithosphere. We have just seen the differences between the meteorite and igneous rocks; furthermore, the presence of chondrules excludes the possibility that the stone was formed under the influence of agents analogous to external geological forces; the complete freshness of the silicate components of the meteorite suggests that the latter is scarcely an ordinary tuffaceous product. It is very probable that this aggregate arose under some special cosmic conditions, different from those that prevail on Earth; it is also possible that, after its formation, the meteorite, during a long wandering in celestial space, underwent rather substantial changes, which could have given it a brecciated appearance.

Among such later changes belongs the fusion crust, which covers the primary surfaces of the specimen and formed at the time when the latter was traversing the Earth’s atmosphere at its inherent cosmic velocity*). This crust is colored carbon-black, here and there with a brownish tint, and has a very weak luster, owing to the locally strong development of various small irregularities on its surface. How characteristic

*) E. Cohen, Meteoritenkunde, II.—Stuttgart, 1903: 94.

for intermediate chondrites, the crust of our meteorite has an insignificant thickness, varying within the range of \(1/4\)--\(3/4\) mm, and is sharply separated from the underlying mass of the stone, which at the very boundary with the crust retains its usual properties. Despite this latter circumstance, the new formation is rather closely connected with the unaltered substance of the meteorite, since the lower boundary of the crust is very finely uneven, and thus a considerable surface of adhesion is obtained between these two component parts of the specimen.

The uneven thickness of the crust is also reflected on the surface of the latter, where there are quite many small tubercles, which correspond to grains of components more resistant to the rapid action of heat that developed during the passage of the meteorite through the Earth’s atmosphere. Such resistant components proved to be pyritic formations, especially schreibersite, which occurs both on the surface and in the body of the crust. Probably, the resistance of the pyritic constituent parts can explain the troilite accumulation, noticeable in places, beneath the fusion crust. Of course, there is no possibility of establishing the influence of the various components of the meteorite on the character of the surface of the crust; one can only further note that the occasionally encountered oval, usually somewhat lustrous spots apparently belong to the more easily fusible chondrules *).

In addition to the tubercles connected with the various grains of the underlying mass of the meteorite, on the surface of the crust there is also a whole series of irregularities of another origin. As was already said above, the meteorite specimen has only two primary (for an aerolite) surfaces with a fusion crust; these surfaces, by the character of the irregularities that are subject to description, differ rather sharply from one another. One surface, by the presence of several gentle depressions, must be assigned to the rear part of the aerolite **), and this surface is covered with very fine, rather difficult-to-distinguish ridges, which form an irregular ripple, as if running away from the strongly rounded edge, free from such irregularities, between the two primary surfaces. Such a position of the ridges undoubtedly indicates that they represent the result of the solidification of molten particles of the meteo-

) A. Brezina.* Bericht über neue oder wenig bekannte Meteoriten.—Sitzbr. d. K. k. Akad. d. Wiss. zu Wien LXXXV, 1882: 339

) W. Haidinger. Eine Leitform der Meteoriten.—Sitzbr. K. k. Akad. d. Wiss. zu Wien. XL, 1860: 532.

…which occurred as a result of strong gusts of air currents, which were produced during the lightning-fast movement of the celestial wanderer through the earth’s atmosphere, as is accepted in the science of meteorites.

Far more original is the other surface of the specimen, having a regular, slightly convex form and probably representing the lateral part of the meteorite when it was piercing the lower horizons of the atmosphere. Here the waves of the moving molten substance solidified considerably more powerfully, increasing the thickness of the crust in places to as much as 1 mm. But the regular structure of these waves proved to have been disturbed at the very end of their formation by gases, as it were, bursting out of the molten mass, and the crust acquired a slightly vesicular texture. And thus, in the depressions of the uneven surface of the crust, we find here and there a rather loose, carbonaceous mass, bound together with dried plant fibers, the continuations of which sometimes adhere firmly to the solidified mass of the crust. There can be no doubt that these fibers belong to the vegetation of that bog into which the meteorite fell; they were seized by the molten substance of the meteorite and partly managed to burn, whereby a certain quantity of gases was evolved; the latter, together with the vapors of the soil water, imparted to the crust the texture indicated above.

This observation, however insignificant it may seem, is quite important in resolving the question of the time at which the fusion crust formed. Indeed, if the solidified waves contain within themselves soil particles of earth, then they must have manifested themselves in the very lowest horizon of the atmosphere; on the other hand, these waves in places seize the entire thickness of the crust. Consequently, the latter formed at the very end of the meteorite’s fall and—almost instantaneously, as has more than once been established for meteorites by various means *)—and as is further confirmed by the sharp boundary between the crust and the underlying mass of the stone. From this one may draw the further conclusion that the principal mass of the molten substance produced during the meteorite’s piercing of the atmosphere was blown away from the latter by air currents, which probably took a substantial part in the luminous phenomena that accompanied the fall of the aerolite. It is only difficult to determine what part of the monolithic piece of the meteorite was torn away in such a manner by the atmosphere, since we cannot yet establish the absolute rate at which the crust formed.

) A. Daubrée. Météorites provenant de la chute qui a eu lieu le 1-er janvier 1869, Hessle, aux environs d’Upsal (Suède). — Comptes rendus*, LXVIII, 1869: 364.

Besides the fusion crust, which is the youngest formation, the present meteorite has yet another macroscopically very conspicuous sign of a later alteration, in the form of so-called black veins. These black veins, the substance of which seems as if identical with the mass of the crust, pass through the body of the meteorite in small number; at least, on the fracture surfaces of the specimen that assumed their final appearance after the separation of pieces of the aggregate, only one such vein is visible, giving off a small branch, while on the fragment of the meteorite used for chemical analysis there were two veins converging at a right angle. The veins examined pass calmly through the grains of the aggregate, do not intersect one another, and are not accompanied by disjunctive displacements, being, apparently, of simultaneous origin. As for the thickness of these little veins, in general it is very small and fluctuates around 0.1 mm. It is characteristic that the only vein visible on the specimen, running approximately parallel to that free surface of the meteorite which we have taken to be the lateral one, becomes noticeably thinner toward the rear part of the stony body; moreover, in this same direction its apophysis separates from the vein. Finally, in order to characterize the formation under consideration it must be noted that the vein does not appear on the fused surface of the specimen, being—therefore—older than the crust.

Proceeding from the described external properties of the veins, we must think that they originated by the same route as the fusion crust; on just this solution of the question nearly all investigators of stony meteorites agree*).

Undoubtedly, the cracks of the veins appeared in the meteorite after it had entered the earth’s atmosphere, and this must be explained not so much by the sudden development of heat and the sharp rise of temperature around the celestial body, as is proven by the existence of the fusion crust, but rather by the mechanical action of the layers of air being pushed apart by the meteorite, whose currents in the present case may be compared with the gusts of gases produced, for example, in the explosion of dynamite**). If this is so, then the air, penetrating with tremendous force even into hairline cracks, could quite easily melt the walls of the latter and partly inject into these cracks the molten—

*) E. Cohen. Meteoritenkunde II.—Stuttgart, 1903; 13°.

**) A. Daubrée. Conséquences à tirer des expériences faites sur l’action des gaz produits par la dynamite, relativement aux météorites et à diverses circonstances de leur arrivée dans l’atmosphère.—Comptes rendus, LXXXV, 1877; 257.

molten substance, which formed in abundance on the surface of the meteorite.

It goes without saying that the strongest action of the atmosphere upon the meteorite that had burst into it occurred at the first impacts, since the subsequent motion of the celestial body, owing to the increasing friction in the denser layers of the air envelope, took place at an ever-decreasing velocity. And this proposition is well confirmed by the fact that the black veins of our meteorite do not protrude on the surface of the stone, being cut off by the crust of fusion which, as was established above, formed at the very end of the meteorite’s fall.

Finally, it may be noted that the veins preserved in the piece of the meteorite arose, nevertheless, in the second half of the phenomenon, for they are few in number and thin, whereas the first cracks broke the primary aerolite into separate fragments, as must be supposed from the strong cracking that accompanied the fall of the Tomsk meteorite.

Let us now turn to the microscope, in order to study more closely the composition and structure of our interesting stone and thereby try to approach the solution of the question of its origin. If ore admixtures are excluded, the composition of the meteorite microscopically, as in the powder, appears very simple: we see almost only one mineral species of silicate, which has a sharp relief with a rather high refractive index and is slightly colored dirty-brown; however, this inconstant coloration of the silicate may—rather—be explained by the phenomenon of pseudochroism, since the grains of the aggregate are distinguished by the presence of an enormous number of cracks of various kinds*). Upon insertion of the analyzer, the picture becomes more complex, but nevertheless the varying interference coloration of the grains in the section can be explained by the different orientation of sections relative to the optical ellipsoid of the corresponding mineral individuals of one species. Only a detailed study of the latter, with measurement of the optical constants, establishes that we are dealing here in reality with several silicates.

This optical investigation was carried out with the application of Fedorov’s universal method, which appears, it seems, to be the first attempt of this kind in the study of meteorites. The Fedorov method, having in general excellent qualities, in the present case proves especially suitable,

*) E. S. Fedorov. Foundations of Petrography.—St. Petersburg, 1897: 83.

because the grains of the aggregate being described contain, as was said above, many small cracks, among which there are also cleavage cracks; and with the aid of the latter it is possible rather easily to find the relation between the optical and crystallographic elements of the crystalline formations, something that only the given method is capable of accomplishing.

Unfortunately, the nature of the material did not allow the microscopic analysis to be carried out with the completeness that is possible for the method chosen, and that would have been extremely desirable in the study of so rare a stone. The fact is that a large part of the aggregate grains have dimensions too small for making precise measurements; furthermore, the meteorite at one time experienced such a substantial shock that its constituent parts ordinarily reveal traces of cataclasis, and this circumstance interferes with the accuracy of optical determinations, whatever method may be applied. In several thin sections of the meteorite, nevertheless, about two dozen such sections of components were found that more or less satisfy the conditions for the proper application of the universal method; the remaining grains of the aggregate were determined by analogy of their external properties with the properties of already known mineral formations.

Thus, according to the investigations, one mineral species of the meteorite is distinguished by comparatively low birefringence, manifested, at normal thickness of the thin sections, in light-gray polarization colors, characteristic, for example, of enstatite or labradorite. Although the universal method does make it possible to determine the magnitude of birefringence rather accurately, I did not attempt to undertake this task, since the cracks by means of which the thickness of the thin section can be measured are here very fine or irregular; as for the relative determiners by which E. Stratanovich*) proposes in general to characterize the birefringence of a biaxial crystalline substance, the small size of the grains and the usually strong cloudy extinction of the latter prevented the precise determination of these constants as well, which have not yet acquired full rights of citizenship. Likewise, the refractive index of this mineral species can be determined only approximately; by the method of setting the principal optical sections, the mean refractive index was found to be about 1.64.

) E. Stratanovich: Relative determiners of the birefringence of a biaxial crystalline substance.—Proceedings of the Mining Institute*, II, 1912: 193.

But let us proceed to a more precise characterization of the mineral. It was easy to establish that the latter has a negative optical sign and a fairly constant angle of the optical axes, fluctuating, in measurements that were not always equally exact, between \(72^\circ\) and \(81^\circ\); as an average, and moreover the most frequently observed value, one may take for \(2V = (-)\ 77^\circ\). The other constants could be obtained by means of the cleavage planes, but, unfortunately, in the mineral grains available for investigation there appears only one system of cleavage, having the character of parting and coinciding with the plane \(N_g N_p\) of the optical ellipsoid.

From the totality of all the data presented, and from the paragenesis of the components of chondrites*) the mineral species investigated must be assigned to the group of bronzite—hypersthene, with an FeO content of \(17\%\)**).

From the preceding it is clear that grains with high birefringence in thin sections, manifesting themselves in red and green polarization colors, belong to another mineral species, which, according to the indicated paragenesis, can in all probability only be olivine. Indeed, these grains have a much smaller number of regular cracks, for which reason they appear somewhat lighter than grains of bronzite, and their principal optical sections are established only on the assumption that \(N_m = 1.68\). In the same way the angle between the optical axes in different measurements fluctuates from \((-)\ 87^\circ\) to \((+)\ 82^\circ\), while the mean and most frequently encountered value is \(2V = (+)\ 88^\circ\). It is characteristic that in almost all sections of this mineral a good cleavage appears, coinciding with the plane \(N_g N_m\) and corresponding to (010) of olivine. There is also a less perfect cleavage along \(N_g N_p\), which plane coincides with (100) of olivine, apparently not having parting in this direction in terrestrial formations***); as for meteoritic olivines, in them such parting has rarely been observed****).

In general, olivines of meteorites differ from the corresponding terrestrial minerals in certain physical properties and, perhaps, in composition; a proposal has even been made to call this mineral species olivine-dome*****). It is difficult to decide to what extent the olivine of our meteo-

*) E. Cohen. Meteoritenkunde. II.—Stuttgart, 1903; 35.
**) E. S. Fedorov. Foundations of Petrography.—St. Petersburg, 1897; Fig. 2, Table II.
***) A. Lacroix. Mineralogie de la France et de ses colonies.—Paris, 1893–5; I, 171.
****) E. Cohen. Meteoritenkunde. II.—Stuttgart, 1903; 276.
*****) Ch. Shepard. Report on American Meteorites.—Am. Journ. of Science and Arts, VI, 1848, 402.

... deserves such identification; for this it would first of all be necessary to carry out partial chemical analyses, but the fine-grained nature of the aggregate and—especially—the close mutual intergrowth of the components of the stone do not permit such work to be performed. As for the physical properties of the olivine of the meteorite, they are indeed somewhat unusual. Thus, macroscopically this mineral cannot be distinguished from pyroxene, whereas in terrestrial rocks it is rather difficult to confuse these two formations. It seems to me, however, that in mountain rocks the sharp difference in the coloration of their constituent minerals is to a considerable degree the result of chemical weathering, which takes place as these rocks, during denudation of the region, approach the earth’s surface, where alone we can study them. It goes without saying that meteorites in general are ideal in their freshness, and therefore silicates analogous in composition have in them approximately the same coloration. Also characteristic of our olivine is the constant presence of cleavage and parting cracks, indicating—though only—that, as a consequence of abrupt changes in physical conditions, the entire mass of the meteorite, as well as its individual components, is in a state of strong internal stress, under which even an insignificant difference in the density of the molecular network along the corresponding possible faces of the mineral is readily manifested.

Rhombic pyroxene and olivine make up almost the entire silicate mass of the olivine. However, one more silicate must be mentioned, small grains of which are occasionally encountered among the other components. These grains have an irregular form, are completely water-clear and almost devoid of any cracks. Optically, this mineral species is characterized by a comparatively low refractive index and very weak birefringence, which nevertheless permits the principal optical sections of the mineral to be established quite precisely; moreover, in one case it was possible to determine that the mineral belongs to the biaxial class, with the angle between the optical axes \(2V = (-)\ 40^\circ\). By all the features listed, this crystalline formation represents the meteoritic mineral of plagioclase composition that bears the name maskelynite*) and which irregularly—

*) N. Winchell. Sur météorite tombée le 9. avril 1894 près de Fischer (Minnesota).—Comptes rendus. CXXII, 1896; 682.

but was taken by Tschermak to be the glass of remelted plagioclase *).

As regards the metallic alloys and sulfides that form the characteristic “admixtures” in the meteorite, their description can be only preliminary, since these components occur, in general, in very small and imperfect crystals and are almost not amenable to mechanical separation; to this it must be added that sufficient methods have not yet been developed for the microscopic study of opaque minerals.

Among the admixtures, the sulfides are the first to catch the eye; for the most part they have a bronze-yellow color in light tones, and by this feature, as well as by their ready solubility in acids without the separation of sulfur, they must be assigned to troilite or to a variety of iron pyrites transitional to magnetic pyrites. Not infrequently this component forms rather considerable accumulations of small grains, or appears in thin, short veinlets; in general, however, the troilite seems to be finely disseminated in the meteorite.

It is not always easy to distinguish from troilite another pyritic component, which has a strong luster with a tin-white color and quite often appears in platy grains of noticeable size. By these characteristics the compound described may be taken for schreibersite, which is, properly speaking, phosphide of iron.

Much less noticeable are equally varied, in size and mode of occurrence, grains and plates of native iron, which, by their prevailing iron-black matte coloration and according to the general chemical analysis of the meteorite, belong to varieties of meteoritic iron containing a little nickel.

Finally, in the chemical analysis of the stone, very small grains and crystals of a peculiar compound were discovered, insoluble in any acids and decomposing only when fused with acid potassium sulfate. According to qualitative determinations, we have here a very indifferent alloy of Ni, Co, Fe, and Cr, with the last predominating; moreover, in this compound a certain amount of S was found. If to what has been said we add that the mineral has a black color with a bluish tint, then the compound must be assigned to the type of the daubréelite group **).

*) G. Tschermak. Die Meteoriten von Shergotty und Gopalpur.—Sitzbr. d. K. K. Ak. d. Wiss. zu Wien. LXV, 1872, 127.

**) E. Cohen. Meteoritenkunde I—Stuttgart, 1894: 211.

A more precise conception of the composition of the Tomsk meteorite will be obtained from consideration of the numerical data, given below, of the chemical analysis of this meteorite. The analysis was performed by a laboratory assistant at the chemical laboratory of the Tomsk Technological Institute, A. P. Kalishev, and, in terms of the number of elements determined, is preliminary; a complete analysis will require the expenditure of a rather large amount of time, and its results, together with a description of the extremely difficult methods of analysis used in this case, will be given by A. P. Kalishev separately in due course.

Chemical composition of the Tomsk meteorite.

\[ \Sigma = 99.56. \]

In addition to the analysis given, it should be noted that the total sum of the last 5 elements was established only approximately, and that the direct determination of FeO encountered, owing to the presence of FeS and Fe, almost insurmountable difficulties, and the content of this oxide was obtained indirectly. Namely, by elutriation (in water, but not in heavy liquids, which react with the substance of the meteorite) it was possible to isolate native iron and thus determine its percentage content in the meteorite; then, in the calculations, the entire amount of S was bound with Fe in the form of FeS, which is fairly close to reality; from these two data and from the total Fe content it was possible to find, with a known degree of approximation, the sought amount of FeO.

To establish from the analytical data the relative content of the various mineral components of the meteorite without partial analyses is impossible, since meteoritic pyroxenes have a very variable composition

and not infrequently contain a noticeable quantity of glass and alkalies*); in view of the latter circumstance, it is not even possible to determine by calculation what quantity of the simplest silicate component in composition—maskelynite—is contained in our meteorite. One can only say that the latter consists approximately of 85% silicate and 15 weight percent ore components.

Let us now turn to consideration of the microstructure of our aggregate. First of all, it should be noted that almost all silicate grains in the thin sections of the specimen are broken up by sharp cracks into very small parts, and in ordinary light one obtains the picture of some kind of irregular—fragmental formation. With crossed nicols, the structure of the aggregate appears somewhat more ordered, since individual adjacent fragments are often connected into an optically homogeneous individual; therefore—among other things—it is quite probable that a significant part of the indicated cracks is evidence of strong stress in the mass of silicates, which arose under special conditions of crystallization of the meteorite substance**), and that these cracks manifested themselves, perhaps, when the stone fell to the surface of the earth.

In addition, in thin sections prepared from those parts of the meteorite where black veinlets are present, we find a very pronounced cataclastic texture, especially clearly manifested when the preparation is rotated about the axes of the universal stage, when almost every grain exhibits strong cloudy extinction. Of course, such a property of the meteorite components cannot be evidence of the manifestation of internal forces, like the part of the fracturing analyzed above; on the other hand, this property could not have been caused by external forces analogous to mountain-forming agents, since nowhere in the thin section are cementation belts observed, so characteristic of terrestrial formations of this kind. If one takes into account that the cataclastic phenomena are found essentially around the black veinlets, then it is natural to recognize for this cataclasis and for the veinlets a common origin.

We have already seen that the black veinlets were the result of strong bursts of air; these explosions must, of course, have caused cataclasis of the meteorite, at least near the cracks. True, attempts have been made to explain cataclastic phenomena, and also the black veinlets themselves, by collisions of the meteorite with other celestial bodies—collisions in which

*) E. Cohen. Meteoritenkunde. I. Stuttgart, 1894: 296.

**) E. Cohen. Meteoritenkunde. I.—Stuttgart, 1894: 325.

in which, undoubtedly, a certain amount of heat is generated). But observations show that no movement in the meteorite was taking place along the cracks filled with black veinlets, and therefore the confinement of the impact heat only to such cracks appears incomprehensible. Furthermore, according to analyses, the substance of the black veinlets proves in general richer in oxygen than the main meteorite*), and this can be explained only by the development of the process that led to the formation of the veinlets, and of the cataclasis itself, in the terrestrial atmosphere.

The described fissuring and, in general, the irregularity in composition give the texture of the meteorite a breccia-like or tuffaceous character. But if we mentally free the pictures given by the thin sections from the first property, then we can find signs of a porphyritic structure, although one differing from the analogous structure of rocks.

Thus, the manner in which the phenocrysts themselves appear is original. Comparatively rarely are they represented by good crystals; for the most part we are dealing with grains of irregular form and of various size, and these grains form very inconstant aggregates, sometimes constituting a significant part of the thin sections, the general character of which varies greatly in different parts of the meteorite.

Furthermore, the pyroxene phenocrysts for the most part have a complex structure, which is expressed in various ways. In some cases the irregularly bounded crystals of this mineral represent quite clear intergrowths of several individuals situated in a sub-parallel relative position, as can be well determined from the magnitude of the angles between the like-named axes of elasticity of the individuals; that such intergrowths are not twins is evident already from the irregular character of the components of the composite crystal. In other cases the individuals not only intergrow along a very broken surface, but also finely penetrate one another, so that not entirely clear crystalline formations are obtained, the true nature of which can be determined only in certain positions of the preparation relative to the plane of symmetry of the microscope; the indefiniteness of these formations is further increased by the circumstance that they usually possess very fine fissuring.

Examining carefully the complex pyroxene phenocrysts, we notice that some of them differ rather—

*) W. Wahl. Beiträge zur Chemie der Meteoriten. Z. f. an. Chemie 69, 1911: 83.

**) E. Cohen. Meteoritenkunde II.—Stuttgart, 1903; 123.

but by regular intergrowth, expressed in the uniform growth of the corresponding individuals when a large number of more or less parallel lamellae are formed. Especially interesting are segregations consisting of two individuals. For such grains of pyroxene it can be established that the relative number of lamellae of the intergrowing individuals varies greatly; somewhere the grains of the given crystalline formation consist almost of a single individual, manifested—however—in a mass of parallel lamellae. It is not difficult to see here an analogy with perthite-albite, in which the substance of the monoclinic feldspar has almost disappeared, while the albite nevertheless retains a finely lamellar structure *).

When the lamellae of such a complex pyroxene individual decrease in size, the pattern of the structure of a chondrule is obtained. And it must be said that some small chondrules have the form of regular polyhedral grains of the given bisilicate. On the other hand, finely lamellar individuals of pyroxene sometimes appear in triangular sections. Thus ordinary chondrules, consisting of finely lamellar sectors, which are arranged often eccentrically and even irregularly, represent formations of the same type as the more or less normal grains.

As regards olivine, the latter always appears in regular grains, often having the character of well-bounded segregations. It does not form independent chondrules in our meteorite, but sometimes penetrates pyroxene chondrules while preserving its granular appearance. Such grains of olivine can only be inclusions in chondrules **).

As has already been noted above, segregations of various kinds are distributed very unevenly in the meteorite—so that different portions of the stone have a somewhat differing mineralogical composition; moreover, they form very irregular accumulations. In view of this, the microscopic picture of the meteorite recalls not so much an igneous rock as a volcanogenic tuff. The resemblance to the latter is further strengthened by the circumstance that the “groundmass” binding the segregations has an irregular grain and differs from these segregations only in the comparatively insignificant dimensions of its components. Despite all these properties of the microscopic structure of the meteorite, it is impossible to agree—

*) M. Usov. Boundary Dzungaria. Vol. II, issue 1. Description of the rocks.—Tomsk, 1911: 405.

**) W. Wahl. Beiträge zur Chemie der Meteoriten.—Z. f. an. Chemie. 69, 1911: 79.

to agree with the view that the intermediate chondrites represent a tuffaceous formation of a type ordinary for our planet, as is accepted by many investigators of meteorites*).

Indeed, with a certain degree of insistence it is not difficult to observe that many grains of the groundmass are fused into segregations or else are intergrown with the latter epigenetically, like the quartz of the groundmass of many porphyry-granites and granite-porphyries); and it must be said that in such intergrowth it is usually the olivine of the groundmass that is encountered. But if the indicated phenomenon does occur, it is absolutely impossible to speak of this chondrite as a mechanical accumulation of tuffaceous particles, even if they were subsequently welded by thermometamorphism*).

Thus, the meteorite described cannot be considered a sedimentary formation, for it consists of components of pyrogenic origin—components which are, in general, very prone to alteration under the influence of gaseous and liquid agents and nevertheless remain remarkably fresh; in the close connection of the grains of the aggregate, which often penetrate one another, the meteorite also differs from simple tuffs. But it also cannot be taken as an analogue of an ordinary eruptive rock, in view of the irregularity of its structure and—especially—the presence of chondrules, which are formations characteristic only of meteorites and which occur together with normal crystals of the same components.

If the analogy under discussion is to be carried through to the end, then the meteorite must also be compared with terrestrial metamorphic rocks. We have already seen that the stone was not subjected to prolonged pressure, and that the observed cataclastic areas appeared as a result of instantaneous explosions in the atmosphere. Then, although skeletal crystals are formed in contact metamorphism, they have nothing in common with chondrules. Finally, in pyrometamorphism the recrystallization of molten material proceeds with the formation of an ordinary texture****), which is not observed in our meteorite. Thus the meteorite described does not represent an analogue of metamorphic formations.

*) G. Tschermak. Die Bildung der Meteoriten und der Vulcanismus.—Sitzbr. d. K. k. Akad. Wiss. zu Wien, LXXI, 1875, 672.

) M. Usov, Petrographic Description. Vol. II, issue 1. Description of rocks.—Tomsk, 1911, 88.

***) W. Wahl. Loco citatio, p. 85.

**) C. Doelter. Petrogenesis.—Braunschweig 1906, 157.

From all that has been said it follows that a meteorite arises under special conditions not encountered on the surface or in the lithosphere of our planet. We can further make only two assumptions concerning the place in which this stony aggregate was formed: it is either the deep part of some celestial body, analogous to our barysphere, which is inaccessible to our investigation, or an airless space within some nebula.

The first assumption must be rejected at once, for at great depths, where in general high temperature and pressure prevail, crystallization proceeds very slowly, and coarse- and uniformly grained dense masses are produced, quite unlike our meteorite. Consequently, the latter could have formed only in the space of an unindividualized nebula.

From the tuff-like structure of the meteorite it is natural to conclude that the latter was formed from separate fragments—just as Chamberlin accepts this in his planetesimal hypothesis.) But such a solution of the question would be insufficient. Indeed, planetesimals, or in general the substance of nebulae, have a low temperature,*) and upon the collision of individual planetesimals not enough heat can develop to bring these bodies into a molten state. Meanwhile, the cosmic substance, during the formation of the mass of our meteorite, was undoubtedly strongly heated, since everywhere we find intimate fusion of the separate components of the stone. Moreover, one must think that many of these components were not yet formed at the moment of their collision, but crystallized during the condensation of the substance into the mass of a chondrite, as is evidenced by the very occurrence of chondrules, which could not have formed in the depths of a large celestial body and therefore could not have been independent planetesimals, representing in general the result of the fragmentation of celestial bodies in cosmic catastrophes.

To what has been said the following addition may be made. We have seen that, in their mode of occurrence and in their relation to the structure of the meteorite, metallic alloys and sulfides do not differ from the silicate components of the stone. And so, if the latter crystallized during the formation of the meteorite, then the same may be said of the indicated “admixtures.” But troilite, as experimental

) Th. Chamberlin & R. Salisbury. Geology. V. II. Earth History.—New York, 1905; 64.
*) Sv. Arrhenius. Das Werden der Welten.—Leipzig, 1908; 175.

studies,*) manifests itself at very high temperatures, while the complex compound of Fe, Ni, Co, and Cr that has been found is in general an indifferent formation in many respects.

Thus, there can hardly be any doubt that the meteorite described, like other analogous celestial stones, was formed from a molten mass that, before solidification, had a high temperature. On the other hand—the nonuniform composition of the meteorite, and the probable formation of the latter in a shapeless nebula, and the presence of chondrules—all this speaks for a dispersed state of the molten mass before its condensation and solidification into an aerolite of one size or another.

W. Wahl, who most clearly expressed the idea of the dispersed state of the molten mass that gives rise to chondrites, believes that the crystallization of droplets into chondrules takes place in a hot atmosphere.) But it seems to us that such a condition is both unnecessary and not feasible in airless space, even one densely populated by planetesimals. The needlessness of the condition advanced by W. Wahl is evident from the fact that chondrules not infrequently contain glass and, as described in the present article, have a structure that is exhibited by silicate melts solidifying rapidly and at low temperature.*)

We now have to clarify the circumstances under which the molten mass that formed our meteorite could then have been dispersed and, finally, rapidly solidified. In resolving this question we shall proceed from the proposition proved above, that the meteorite formed in airless space. Bearing this in mind, we can in no way imagine an aerolite as a normally accumulated aggregate of molten planetesimals. Thus, the melting of cosmic matter, its dispersion, and its subsequent coalescence back into the compact mass of a meteorite—all these processes took place very rapidly and were separated by insignificant intervals of time.

There can hardly be any doubt that such a change of phenomena, and even the very formation of molten matter, takes place in cosmic catastrophes caused by the collision of enormous cosmic masses moving at astonishing speed. In such a collision or

*) E. Man, I. Creashaw, J. Johnston, and E. Larsen. Die mineralischen Eisensulfide.—Z. f. an. Ch. 76, 1912: 273.

**) W. Wahl. Beiträge zur Chemie der Meteoriten.—Z. f. an. Chemie. 69, 1911; 80.

***) M. Usov. On metasilicates of migration and iron.—Izv. SPb. Politekhn. In-ta, XIX, 1913: 425.

even with a close passage of celestial bodies*), at least one of them breaks up into fragments of the most varied size, with the greater part of the latter being temporarily in a molten state; some of these fragments have melted under the influence of the heat developed by the impact, while others—and, probably, the predominant ones—come from the inner parts of the celestial body, which even during the lifetime of the latter had a very high temperature.

All these fragments and molten particles, of the most varied size, fly apart in definite directions, repeatedly colliding with one another. It is quite natural that in such collisions individual liquid or plastic particles weld together with one another and with solid fragments, and at the same time undergo more or less rapid crystallization, quickly entering cold cosmic space. In this way irregular bodies of diverse sizes and structure may be produced, subsequently entering into the composition of a nebula and constituting a considerable part of the planetesimals.

The theory being set forth seems to explain rather well both the irregular, tuff-like composition of chondrites, consisting of crystalline elements of various origin, and the formation of chondrules, as rapidly crystallized droplets of a supercooled silicate melt**), as well as the constant manifestation of fusion among many components of the stony aggregate, and the acquisition by the latter of strong internal tension. In the same way it is easy to understand the considerable looseness of some stony meteorites, as a consequence of the fact that these bodies were composed of solid fragments and of already viscous, sufficiently cooled, molten particles.

Thus the chondrites that reach the surface of the Earth, constituting the principal mass of stony meteorites, are direct witnesses of the formation of nebulae, chiefly of the solar nebula. And the Tomsk meteorite long wandered in airless space, and only now, having fallen upon the surface of our planet, has it found temporary repose.

*) Th. Chamberlin and R. Salisbury, Geologie, V. II, Earth History.—New York, 1906; 51.

**) W. Wahl, loco citato: S. 78.

Tomsk, April, 1915.

M. Usov.