Тунгусский феномен / Исследования / Библиография / 1990-99 / 1994 / G. Longo, R. Serra, S. Cecchini and M. Galli, SEARCH FOR MICROREMNANTS OF THE TUNGUSKA COSMIC BODY Planetary and Space Science 42, n.2, p. 163-177; 1994

G. Longo, R. Serra, S. Cecchini and M. Galli, SEARCH FOR MICROREMNANTS OF THE TUNGUSKA COSMIC BODY
Planetary and Space Science 42, n.2, p. 163-177; 1994

SEARCH FOR MICROREMNANTS
OF THE TUNGUSKA COSMIC BODY

G. Longo^1-2, R. Serra¹, S. Cecchini³ and M. Galli¹

1 - Dipartimento di Fisica dell'Universita' di Bologna, Via Irnerio 46, Bologna (Italy)
2 - Istituto Nazionale di Fisica Nucleare, Sezione di Bologna, Via Irnerio 46, Bologna (Italy)
3 - Istituto di Studio e Tecnologie delle Radiazioni Extraterrestri, CNR, Via De' Castagnoli 1, Bologna (Italy)

(Published on: Planetary and Space Science 42, n.2, p. 163-177; 1994)

Abstract
A new method is here used to obtain experimental data on the nature and composition of the Tunguska Cosmic Body. The leading hypothesis of the method is that in the forest conifers which have survived the Tunguska catastrophe, the fluid resin present at the moment of the event could have acted as a trap for airborne particles, as happens in amber. The growth rings of trees can give information on the age of the resin and therefore on the time when the particulate was trapped in them. With a scanning electron microscope, a total of 7163 particles were found in resin-samples from Tunguska branches and from two control trees. The time distributions of the particles in the years 1885-1930 show clear abundance peaks centred on 1908. This makes it possible to conclude that the presence of Fe, Ca, Al, Si, Au, Cu, S, Zn, Cr, Ba, Ti, Ni, C and O in the particles observed is related to the Tunguska event and that these elements are probable constituents of the Tunguska Cosmic Body. This result is compatible with recent calculations showing that the Tunguska body can be a normal density stony asteroid.

Introduction
On 30 June 1908 the explosion in the atmosphere of a cosmic body over the basin of the Podkamennaja Tunguska river (Central Siberia) resulted in 2,150 km² of forest fall and a great number of trees and branches burnt in a large part of the explosion area.
Since 1927 a great number of scientific expeditions has worked in Tunguska to collect data on the effects of the explosion and search for fragments or traces of the material from the disrupted Tunguska Cosmic Body (TCB). Some reviews summarize the results acquired by these intensive investigations in successive periods, i. e., 1927-1958 (Krinov 1966), 1958-1969 (Vasiljev 1984), 1970-1980 (Vasiljev 1986), 1980-1985 (Vasiljev 1988). Data on forest devastation and records of the atmospheric and seismic waves have made it possible to deduce the main characteristics of the Tunguska explosion, i. e. its exact time, 00^h 14^m 28^s UT, (Ben-Menahem 1975), the coordinates of the point usually called "epicentre", 60° 53' 09"; N, 101° 53' 40"; E, (Fast 1967), the energy release and height of the explosion, though the values for the last two parameters are estimated with great uncertainty. However, no impact craters or meteorite fragments have ever been found in an area of 15,000 km² (Vasiljev et al. 1990), so that the nature and composition of the TCB are still controversial.
Apart from less conventional hypotheses involving near critical fissionable material (Zigel' 1959, Hunt et al. 1960), antimatter meteors (Cowan et al. 1965), tiny black holes (Jackson and Ryan 1973) or alien spacecrafts (Kazantsev 1946, Baxter and Atkins 1976), the two most plausible explanations consider the explosion in the atmosphere at an altitude of 5-10 km of a comet (see, e. g., Whipple 1930, Kresak 1978) or an asteroid-like meteorite (see, e. g., Kulik 1939, Sekanina 1983).
In the recent past the cometary hypothesis has been favoured on the basis that a low density object was needed to explain the Tunguska catastrophe (Petrov and Stulov 1975, Turco et al. 1982). Subsequently, to account for the concentration of energy release of the explosion, two sub-versions of this hypothesis have been developed, one introducing chemical reactions (Tsymbal and Shnitke 1986), the other nuclear-fusion reactions (D'Alessio and Harms 1989).
On the other hand, it has been shown (Grigoryan 1976, Grigoryan 1979, Passey and Melosh 1980, Levin and Bronshten 1986) that the fragmentation of a normal density object can greatly increase the rate of energy deposition in a small region near the end of the trajectory, thus appearing as an atmospheric explosion. Detailed calculations which include the effect of aerodynamic forces that can fracture the object, and the heating of the bolide due to friction with the atmosphere, have recently been performed, showing that the Tunguska event is fully compatible with the catastrophic disruption of a 60-100 m diameter asteroid of the common stony class (Chyba et al. 1993, Hills and Goda 1993). However, due to the uncertainty of such input parameters as the energy and height of the explosion or the inclination angle and the encounter velocity of the impactor, the same calculations do not exclude the possibility that the TCB was a high velocity iron object, nor rule out a carbonaceous asteroid as an explanation of the event.
The only way to achieve certainty about the nature and composition of the TCB remains the search for some of its remnants. Many radiocarbon analyses of Tunguska wood samples (Nesvetajlo and Kovaliukh 1983), chemical analyses of soil or plants, bed-by-bed chemical analyses of the peat formed by Sphagnum fuscum in 1850-1950 (Vasiljev et al. 1973, Golenetskij et al. 1977a, Golenetskij et al. 1977b, Kolesnikov et al. 1977), isotopic analyses of many different soil, peat and wood samples (Kolesnikov et al. 1979), as well as analyses of the spherules from Tunguska soil samples collected in a radius of several tens of kilometres from the epicentre (Jéhanno et al. 1989, Nazarov et al. 1990), have been carried out. Nevertheless, many conclusions of this intensive work are still uncertain, so that further investigations are needed.
In the present work a new method is used to obtain experimental data on the nature and composition of the TCB. Preliminary attempts (Cecchini et al. 1992, Valdré and Korlevic' 1993) to use this method have shown its validity, although the necessary corrections (Galli et al. 1993) had to be introduced in order to deduce a reliable dating for the particles.

Method
The leading hypothesis of the method is that in the forest conifers which have survived the Tunguska catastrophe, the fluid resin freshly emitted due to some physiological effect, could have acted as a trap as happens in amber, for airborne particles present at the moment of the event. The most interesting case is when the resin is emitted and deposited around small dead branches, naturally decorticated long before the event and subsequently merged in the growing stem. The resin produced inside the sapwood is partially emitted year by year to the exterior, in a fluid state, during the most active vegetation period. The part of it immediately surrounding the branch on the outer bark of the tree, solidifies and becomes gradually included from outside, by the growing wood. So, the fresh resin can act like a trap for airborne microparticles and part of them, embedded in the tree-resin, can eventually be found in the resin layer along the limit between the growth rings and the withered branch. Then, the elemental composition of the particles can be determined by the analytical electron microscopy technique.
The growth rings, being rather easily datable year by year, can give information on the age of the resin and therefore on the time when the particulate was trapped in it. To date the resin, the width of the outside bark and the width of the resin layer on the bark must be taken into account because they are responsible for a certain shift between the date of the particle capture and that of the adjacent tree rings, as can be seen in Fig. 1.

Fig. 1. Sketch of the inclusion mechanism of airborne particles in the resin emitted around a dead branch of a living conifer.

The sketch in Fig. 1 shows a vertical section of a stem slice containing a withered branch. In the upper part of the figure, the slice is represented in the 1908 state with its last growth ring adjacent to the bark. In the vegetative period of that year, the fresh resin surrounding the dead branch on the outer part of the bark could trap airborne particles. In the lower part of the figure the same section is shown some years later. The solidified 1908-resin with the trapped particles has been included inside the growing tree and now has become adjacent to the growth ring of 1914 (or to a couple of rings depending on the relative thicknesses of rings and resin layers). The shift between the age of the resin and that of the tree-rings varies for different trees and should be experimentally determined.
The correction for the bark and resin widths was overlooked in the preliminary attempts to search for microremnants of the TCB in the tree resin (Cecchini et al. 1992, Valdré and Korlevic 1993) therefore the particles listed in these preliminary works, having been trapped in the resin some years before 1908, cannot be related to the Tunguska event. The refinements of the method, suggested by M. Galli (Galli et al. 1993), have permitted the present systematic analysis and the correct dating of the particles embedded in the tree-resin before, during and after the Tunguska event.
Previous chemical analyses of materials from Tunguska have generally been carried out by searching for traces of elements from the TCB in soil, plants and rocks (Kovalevskij et al. 1963, Iljina et al. 1971, Golenetskij et al. 1977a), in beds of peat (Golenetskij et al. 1977a, Golenetskij et al. 1977b), in growth rings of trees, etc... These analyses - in which infinitesimal mass concentrations of some elements, possibly originated from the products of the explosion of the TCB, were sought in the whole mass of the materials considered - have given uncertain results. Attempts to consider spheroidal particles found in the soil as relic candidates for the TCB came up against the difficulty of reliably dating the spherules. A careful comparison of their composition with that of spherules of known origin has led to the conclusion that about 95 % of the spherules examined have a terrestrial or micrometeoritic origin, and the difficulty of dating the remaining part left uncertain their provenance from the TCB (Jéhanno et al. 1989, Nazarov et al. 1990). The great advantages of the new method here used are that it gives the possibility to bring out and analyse particles which could be pure microremnants of the TCB, and to reliably date them. The same advantages should reveal themselves also when the method is applied to investigate the isotopic composition of the elements found.
The refined method can in principle be applied to get information on environmental conditions in the past, especially in connection with events of microsized particulate production (aerosols from human, volcanic and seismic activity, forest fires, etc...). The accumulation in the resin can be thought to be similar to the deposition and accumulation of volcanic aerosols and particulate in polar ice sheets. So, for example, in the case of volcanic eruptions, large quantities of aerosols and ashes are injected into the atmosphere and are generally deposited within 1 or 2 years so that any "clean-air" location, even at a great distance from the site of the event, must be expected to receive significant amounts of fall-out, which can be subtracted from the background and analysed in a way similar to the present one.

Samples
To collect the necessary wood samples from living trees that have survived the catastrophe, the authors of the present work participated, in the summer of 1991, in a scientific expedition (Cecchini et al. 1992, Galli et al. 1993) led by N. V. Vasiljev, deputy chairman of the Commission on Meteorites and Cosmic Dust of the Siberian Section of the Russian Academy of Sciences, and by G. V. Andreev from the Astronomical Observatory of Tomsk.
Among the few conifers per hectare that have survived the Tunguska catastrophe in the vicinity of the epicentre, the most frequent are Siberian larches (Larix sibirica) and common pine (Pinus sylvestris) followed by Siberian spruce (Picea obovata) and Siberian pine (Pinus cembra). The most suitable tree for our purpose turned out to be the Siberian spruce, with abundant resin and very regular yearly lignification cycles (i. e. without double or missing annual rings), so that the latewood tree ring limits could easily be dated. Therefore, the samples have been taken from trees of this species plus, for comparison, some samples from a Siberian pine.
The samples collected have the form of thin transversal stem slices, sawn at different heights, or of cylindrical cores extracted from living trees and containing the embedded withered branches surrounded by resin.
The samples of interest contained withered branches, dead at least some years before the catastrophe of 1908. On such branches the living tissues have deposited resin both from years before and after the catastrophe. The traces of the withered branches could be discovered by looking in the bark of the stem for spots of shiny fresh resin, the oldest dead branches being more abundant in the lower part of the stem because of the natural pruning of the trees in a normally thick forest. The greater part of the samples was extracted directly from living trees with a 22 mm diameter corer. In order to verify the adequate age of the trees, preliminary cores were extracted with a 5 mm Pressler probe. The wounds due to the perforations were immediately treated with an appropriate cream.
The wood samples collected in Tunguska were subsequently examined in Bologna. In order to bring out the branches, the branch-stem interstices, containing the resin layers, were opened. It was then possible to proceed to electron microscopy inspection of the samples in two different ways: a) through scanning of the surface of the resin that was adjacent to the dead branches; b) by direct observation of the particles extracted after dissolving the resin inside filtering capsules placed in a Soxhlet extractor.
The overall size, morphology and chemical composition of the particles have been determined using a scanning electron microscope equipped with an energy dispersive X-ray spectrometer (in the present work a Philips SEM 515, with EDAX PV9100). A computerised electron detector has been used to determine the concentration of components with atomic numbers greater than 10. The samples extracted from the surface of the resin have been placed on standard SEM stubs and coated with thin films of carbon by evaporation in vacuum.
The resin deposited in the years ¸ 1885-1930 on 14 branches of 7 trees situated in different directions within a radius of 8 km from the epicentre of the Tunguska catastrophe, has been examined (see Table 1). The exact location of the trees listed in Table 1 is given elsewhere (Serra et al. 1994). For comparison, 6 other branches of a control tree growing at about 1,100 km W-SW from the Tunguska site and the resin deposited on the roots of a control tree uprooted by the explosion of 1908 have also been examined.

Results
In the 71 resin samples from 20 branches and 3 roots of the 9 trees listed in Table 1, 7163 particles were found up to 1 June 1993. From this total, 171 pure gold particles, mainly found in the resin of the nineteenth century, clearly having nothing to do with the Tunguska event, will be discussed elsewhere. Here the other 5854 particles found in the resin-samples from Tunguska branches are discussed and compared with the 1138 particles from the two control trees.
The shift between the age of the resin and that of the tree rings has been determined by measuring the bark-width and the width of the tree-rings in all the wood-samples considered, taking into consideration the bark-width and the thickness of resin layers on living younger trees. This shift has been found equal to 6 ± 3 years. The dating given hereafter is referred to the resin age, calculated with a 6-year shift with respect to the adjacent tree-rings. Taking into account the uncertainty in the resin dating, three periods are here considered: 1902-1914, which corresponds to particles that could be related to the Tunguska event; 1885-1901 and 1915-1930, which correspond to "background particles" trapped before and after the Tunguska event.
As mentioned above, a chemical semiquantitative analysis of the composition of the particles has been performed by using a windowed energy dispersive X-ray spectrometer. Great care has been devoted to avoid contamination of the samples during their preparation and analysis. The series from the same branch of the resin samples corresponding to the three periods considered were carbon-coated contemporarily. To exclude artifacts, for six branches, two or three different series of resin-samples for each branch have been prepared at different times after a general check of the experimental set-up.
The analysis has revealed the presence of 28 elements in the particles from the Tunguska tree-resin. In Table 2, the elements present with more than 1 % in at least 1 particle are listed, taking into account that the equipment used cannot reliably detect quantities < 1 % or elements with atomic number Z < 10. These elements are divided into two groups. A first group of elements, which is present in all three periods considered is here indicated as "low-Z" elements. It corresponds to elements with atomic number Z < 20, plus iron. A second group of elements, mainly present in the period 1902-1914, is here called "high-Z" and contains elements with Z >= 22 with the exclusion of iron.
The 5854 particles found in the resin samples from Tunguska branches are split in Table 3 into "low-Z" particles, i. e. particles containing only low-Z elements, and "high-Z" particles, which contain at least 1 % of 1 high-Z element (the high-Z particles usually contain a certain percentage of low-Z elements). As can be seen from the data in Table 3, one third (about 130 particles/cm²) of the "low-Z" particles could be related to the Tunguska event, but it is difficult to distinguish them from the two thirds (about 250 particles/cm²) of "background" particles. Therefore, the analysis is hereafter focused on the high-Z particles, whose density is increased by more than a factor 10 in the period 1902-1914.

Figure 2

Fig. 2. Partition of the 5854 particles found in the resin samples from Tunguska branches for the three periods considered.

The bar graph in Fig. 2 summarises the total number of high-Z and low-Z particles found in the three periods considered. Due to the fact that for some branches the resin was available only for one or two of the three periods of interest, the total number of particles observed is different in the three periods. That is why in the following graphs the results are presented in per mil calculated with respect to the 1181, 3356 and 1317 total particles found in the corresponding period.
As can be seen from Fig. 3, a clear abundance peak is found in correspondence with the period of the Tunguska event. Taking into account the ± 3 years uncertainty in the dating, the particles under the whole peak can be considered related to the Tunguska event, after subtraction of a very small "background". Similar peaks have separately been obtained for the different trees examined (Serra et al. 1994).

Figure 3

Fig. 3. Time distribution (per mil) of the 518 high-Z particles found in the resin samples from Tunguska branches.

The graphs of Fig. 4 show the time distribution of particles containing a single high-Z element, such as copper, chromium, barium and gold (the last only when mixed with other elements). Similar peaks appear also for the other high-Z elements for which sufficient statistics are available, i. e. zinc, titanium and nickel.

Figure 4

Fig. 4. Time distribution (per mil) of particles containing Cu, Cr, Ba and mixed-Au.

As mentioned, 471 out of the 518 high-Z particles contain a certain percentage of low-Z elements. For these mixed particles, considered either in their whole or separately for calcium, iron, silicon, alluminium, sulphur, potassium and chlorine, mixed with high-Z elements, clear peaks centred at 1908 are obtained, as shown in Fig. 5 for the first 4 cases.
The high-Z particles found in the 1902-1914 resin from Tunguska branches, appear to have experienced heating and melting. Many of them are spherules or rounded particles like those shown in the micrographs of Fig. 6.

Figure 5

Fig. 5. Time distribution (per mil) of particles containing a mixture of low-Z and high-Z elements and of particles containing Ca, Fe, Si mixed with high-Z elements.

Figure 6

Fig. 6. Micrographs of four high-Z particles found in the Tunguska branch-resin:
a) a spheroidal particle containing P (34%), Ca (30%), K (18%), Fe (6%), Zn (4%), Ba (4%) and Si (3%) (tree n° 7, year 1904),
b) a rounded particle containing Fe (77%), W (7%), Al (4%), Cl (3%), K (2%), S (2%), Zn (2%), Cu (1%) and Mn (1%) (tree n° 1, year 1910),
c) a particle containing Ti (75%), Ca (14%), Si (9%) and K (2%) (tree n° 7, year 1904),
d) a particle containing Al (28%), Zn (23%), Ca (17%), Si (13%), Fe (10%), S (3%), K (2%), Cl (2%) and Ti (1%) (tree n° 6, year 1907).

Figure 7

Fig. 7. Micrographs of two low-Z particles found in the Tunguska branch-resin:
upper) a particle containing Ca (65%) and Si (35%) (tree n° 1, year 1925),
lower) a particle containing Si (77%), Ca (9%), Fe (7%), Al (4%) and K (2%) (tree n° 6, year 1906).

Among them sharp-edged or flocculent particles are generally absent, unlike what happens for part of the low-Z particles trapped in the resin of the three periods considered (see Fig. 7). The dimensions of the particles can be estimated from the scale bars on the micrographs.
The spheroidal shape is a surely distinctive feature of particles which underwent a thermal effect. Out of the 5854 particles from the Tunguska branches, 88 are spheroidal (77 spheroids are high-Z particles and 11 are pure low-Z particles). The time distribution for the 88 spheroidal particles as well as that for 54 spheroids containing copper, show the usual peaks in correspondence with the Tunguska event, as shown in Fig. 8. Similar graphs are obtained also for the 31 and 30 spheroids containing aluminium and calcium, respectively. It should be noted that the products of a high temperature fusion followed by a rapid cooling are very different one from another, as was experimentally checked by melting meteorite fragments in an electric arc. For example, some spherules obtained in these conditions contain a very high percentage of nickel, while this element is completely absent in others (Florenskij et al. 1968).
On the basis of the chemical semiquantitative analysis performed, the "average composition" ( % ) of the 463 high-Z particles of the period 1902-1914 has been calculated and is shown by black bars in Fig. 9. The quotes mean that the "average composition" should be considered only as a qualitative indication. Indeed, as previously pointed out, the equipment used cannot reliably detect elements with Z < 10 or quantities < 1 %. Furthermore, the measured composition depends on the place where the electron beam hits the particle and on the angle between the electron beam and the surface of the particle. In averaging the composition, the masses of the particles have not been taken into account. Finally, this "average composition" refers to high-Z particles only. Taking into account that probably about one third (see Table 3) of the "pure" low-Z particles has the same origin as the high-Z particles, the percentage of low-Z elements should be increased. It is worthwhile to note, however, that this qualitative composition clearly differs from the composition of the 79 high-Z particles found in the root-resin of control tree n° 2 and of the 29 high-Z particles found in 1902-1930 in control tree n° 10 (see Fig. 9).

Figure 8

Fig. 8.
Upper: time distribution (per mil) of all spheroidal particles.
Lower: time distribution (per mil) of spheroidal particles containing Cu.

After completion of the present analysis, some measurements have been performed with a new scanning electron microscope, a Philips PXL30 equipped with an EDAX PV9900 spectrometer. With the equipment now available, it has been possible to push the analysis down to elements with Z = 5. Two spectra of newly examined particles and the corresponding micrographs, are given in Fig. 10. As can be seen, these spectra clearly reveal the lines of oxygen. To search for the presence of carbon, some resin-samples have been coated with gold instead of carbon. Thus, it has been possible to ascertain that many elements listed in Table 2 are present not only in oxides, but also in carbonates. This can strongly alter the measured ratio between the different components of the particles. Thus, for example, due to the presence of oxygen, the copper/aluminium ratio in the spherule of Fig. 10 is increased by a factor of 1.5 when obtained by using the new equipment with respect to the data obtained in the present work.

Figure 10

Fig. 10. The spectra and the corresponding micrographs obtained with an EDAX PV9900 spectrometer for two particles found in the 1908 resin of tree n° 5.
Left: a spheroidal high-Z particle containing oxides of copper (86.6 %) and aluminium (13.4 %).
Right: a rounded low-Z particle containing oxides of Si (67%), Al (24%), Na (5%), K (3%) and Fe (1%).

Figure 11

Fig. 11. The measured radiocarbon abundance in percent of the modern concentration (pMC) as a function of the tree-ring date for wood samples from tree n. 1. The points connected with a solid line correspond to the data from Stuiver and Becker 1993.

Taking into account the results summarised in the present Section, one can conclude that there are 14 elements, listed in Table 4, whose presence is related to the Tunguska event. The list in Table 4 is given in decreasing order of importance for the first 12 elements. The decreasing order takes into account the number of particles, their average composition and the probable relation to the Tunguska event of part of the "pure" low-Z particles above the background. Oxygen and carbon, whose presence has been ascertained, but for which no quantitative data are available, have been added to the list. The elements Cl, K, Na, Pb and Mn, though listed in Table 2, are not included in Table 4, their relation to the Tunguska event being dubious, due to the low overall percentage of the first three elements and to peaks not so clear as those of Figs. 3-5 for Pb and Mn. The rather poor statistics for Mg, P, Sn, Sr, Br, Ag, Bi, Co, Sb, Cd, W (from 11 to 1 particles, in decreasing order) has led to the exclusion of these elements from Table 4.
As pointed out above, the present method should reveal its advantages also in carrying out the isotopic analysis of the particles found. Up to now, only the traditional radiocarbon method, which consists in searching for infinitesimal concentrations of ¹⁴C in the full mass of the tree-ring material, has been applied to study the Tunguska wood. The radiocarbon analysis of a set of our wood samples from tree n. 1 has been carried out by R. P. Beukens at the IsoTrace Radiocarbon Laboratory of the Accelerator Mass Spectrometry Facility at the University of Toronto (Canada). Figure 11 shows the measured values, with their error bars, of the radiocarbon abundance in percent of the modern (year 1950 AD) concentration, as a function of the ring date. These results are in general agreement with those from Stuiver and Becker 1993. Radiocarbon due to neutron production in a possible cometary impact producing a fusion reaction that might affect the radiocarbon concentration at ground can be ruled out. The graph suggests an increase of radiocarbon in the late 1909 and early 1910. A similar increase, which has probably nothing to do with the Tunguska event, has been found in 1908-1910 tree-rings in previous radiocarbon analyses of Tunguska wood samples (Nesvetajlo and Kovaliukh 1983). This increase may be related with the expected solar cycle modulation of the cosmic radiation (Attolini et al. 1989), though the large experimental error in the 1910 point leaves doubts on such a conclusion.

Discussion
The sudden increase in the resin samples from Tunguska branches of particles containing the elements listed in Table 4 cannot be related to volcanic eruptions or tectonic activity. The dates and places of such events do not correspond to the present data and the morphology of the particles observed is different.
None of the known volcanic eruptions (Simkin et al. 1981) can explain the peaks shown in Figs. 3-5 and 8. In the period of interest the eruptions that have significantly affected light transmission in the northern hemisphere with a certain observed dust veil are those of the volcanoes Soufriére in West Indies (1902), Santa Maria in Guatemala (1902), Ksudach in Kamchatka (1907) and Katmai in Alaska (1912). The last is apparently the only one after 1908 able to cause a noticeable decrease in atmospheric light transmission (Hoyt and Frölich 1983, Lamb 1988). On the other hand, the study of the content of Antarctic snow (Boutron 1980) has shown that the strong peaks observed in the concentration of Zn, Pb, Cu and Cd, appear to parallel rather well the variations of global volcanic activity. In the present work, no peaks of particles containing these elements have been observed in the years 1902-1904 or 1912-1914, which might correspond to three of the four volcanic eruptions mentioned. An increased concentration of the same metals in snow and plants has been observed after earthquakes and it has been shown that one of the most important sources of metals in the atmosphere are aerosols which are emitted during degassing in active tectonic zones (Alekseev and Alekseeva 1992). To our knowledge there is no record of a particular seismic activity that can explain the peaks centred at 1908, observed in the present work. Therefore, the volcanic and tectonic activity can account for the presence in the branch-resin of some "background" particles only.
Another source of particles in the atmosphere is the heavy metal release from plants (Beauford et al. 1975, Beauford et al. 1977), which can obviously be excluded as an explanation of the time-distribution data found for the particles from Tunguska branch-resin.
Nor can the extraterrestrial dust particles (see, e. g., Sandford 1987, Bradley et al. 1988) explain the time distribution of the particles observed, though they are likely to be responsible for part of the "background" of about 250 low-Z particles/cm² and 5 high-Z particles/cm² found in the Tunguska branch-resin in the three periods considered (see Table 3). Anthropogenic sources of aerosols cannot explain the Tunguska data, though they may be responsible for the hint of an increase of high-Z particles observed in the control tree n° 10 (40 km from Tomsk) in the second and third decades of the present century. These conclusions are confirmed by the results of the investigation of heavy metal deposition in the peat of the basins of Podkamennaja and Nizhnjaja Tunguska rivers (Bojarkina et al. 1986). A bed-by-bed neutron activation analysis of the peat from different places located more than 80 km from the smallest inhabited centres and more than 700 km from the nearest industrial cities (Krasnojarsk, Irkutsk and Norilsk) was performed. The yearly deposition per square meter of heavy metals was found more or less constant for the period 1765-1972 and equal to 4000 ± 1000 µg for iron, 5 ± 1 µg for scandium, 3 ± 1 µg for cobalt and 1.2 ± 0.2 µg for gold. These values, therefore, correspond to the deposition in the Tunguska region from all natural sources and can be considered as an upper limit for extraterrestrial dust deposition. These measurements do not provide information about the Tunguska event due to the fact that the yearly deposition was deduced from the content of finite thickness bed peats, one of which corresponds to the whole period from 1891 up to 1910. Control measurements performed by the same authors near an industrial centre showed an increase in the deposition from 1903 to 1978 by factors of about 19 (Fe), 7 (Co and Sc) and 5 (Au).
The foregoing considerations lead to the conclusion that the increase in 1908 of particles containing the elements listed in Table 4 is directly correlated to the Tunguska event. The data obtained from the Tomsk control tree n° 10, not showing such an increase, favour this interpretation. It is unlikely that the particles observed correspond to the part of the material of the TCB that was lifted into the higher atmosphere and precipitated worldwide in the following months. This material was dispersed over a very large area and there is no reason for a particular intensity of the precipitations in the vicinity of the epicentre.
A second possible relation with the Tunguska event is that the particles observed were transported from the ground to the tree-resin by the blast wave from the explosion. A strong argument against this hypothesis is the morphology of the particles observed in the peaks centred at 1908. As shown in Fig. 6, the particles have no sharp edges and show evidence of drastic heating, unlike what one would expect for particles coming from the ground. Moreover, it is difficult to see a relation between the list of Table 4 and the numerous chemical analyses of Tunguska soil, plant and rock samples carried out by different groups (see, e. g., Kovalevskij et al. 1963, Iljina et al. 1971, Golenetskij et al. 1977a). Finally, if the particles were transported from the ground, a similar composition would have been found for the particles trapped by the resin on the roots of control tree n° 2 and those on the branch-resin, but this does not happen, as Figure 9 shows. Though this graph should be considered with caution due to the fact that the roots examined pertain to one tree only, it favours the hypothesis of a different provenance for the particles embedded in the resin of the roots and those from branch-resin. Indeed, at the moment of the explosion, the resin on the branches was ready to receive the particles, while the roots were under the ground. The roots emerged and began to emit resin only several tens of seconds later. Due to this time delay the root-resin could not receive particles directly from the exploded body but was able to trap particles lifted from the ground.
Though a provenance from the ground of part of the trapped particles cannot be excluded, it seems reasonable to assume that at least the main part of the particles above the background came directly to the branch resin inside the blast wave generated by the explosion of the TCB. This conclusion is corroborated by the fact that the particles show evidence of heating and melting as it should be, taking into account that the shocked gases, concentrated in front of the impacting body, attained a temperature of 25,000-30,000 K (Chyba et al. 1993). Therefore, the elements listed in Table 4 are probable constituents of the TCB and the present results can be considered as experimental evidence in favour of the hypothesis of the disruption of an ordinary stony meteorite.
An isotopic analysis of some elements present in the particles found (whose masses are generally equal to fractions of a nanogram) would be of great value in order to distinguish between a cosmic or terrestrial origin of the particles. This isotopic analysis can also help to definitely answer the question of nuclear processes invoked by some authors to explain the Tunguska explosion. No trace of a nuclear process has been found by using the traditional radiocarbon method applied to the full mass of the tree-ring material (see previous Section) and no convincing evidence in favour of such a hypothesis has ever been found (see, e. g., Kirichenko and Grechushkina 1963, Emeljanov 1963). It should be noted, however, that the methods used up to now cannot be considered conclusive since they generally deal with radioactivity measurements carried out more than half a century after the event or with the search for infinitesimal isotopic anomalies in the whole mass of tree-rings, sample of peat, etc..., resulting in measures at the limit of experimental possibilities.

Conclusions
The branch-resin method here used has shown its validity in investigating past events accompanied by aerosol production. Its application to the study of the Tunguska event has made it possible to recognise a first list of 14 elements as probable constituents of the TCB. This list of elements is fully compatible with recent calculations (Chyba et al. 1993, Hills and Goda 1993), showing that the TCB can be a normal density meteorite. Isotopic analysis data together with an extension of the chemical analysis would be useful to confirm the present conclusions.

Acknowledgements
The authors gratefully acknowledge G. V. Andreev, E. M. Kolesnikov, V. D. Nesvetajlo and N. V. Vasiljev for their help in Tunguska and useful discussions, K. Korlevic' for sending to M. Galli the first Tunguska wood sample, whose structure examination excited the idea of the branch-resin method here used, G. Valdré for having carried out a preliminary search of particles trapped in the resin of the same sample and R. P. Beukens for the radiocarbon analysis of a set of our Tunguska wood samples. The authors are indebted, for their assistance in using different electron microscopy facilities, to G. Tarroni and T. De Zaiacomo from the Bologna Centre of ENEA, to F. Sandrolini and A. Saccani from the Department of Engineering of the University of Bologna and to A. Garulli, F. Corticelli and D. Govoni from the LAMEL laboratory of CNR in Bologna.

References
Alekseev, V. A. and N. G. Alekseeva, Investigation of metal transfer in the biosphere during gaseous emission in zones of tectonic activity using methods of nuclear-physics, Nucl. Geophys. 6, n° 1, 99-110, 1992.

Attolini, M. R., M. Galli, T. Nanni and P. Povinec, A cyclogram analysis of the Bratislava ¹⁴C tree-ring record during the last century, Radiocarbon 31, n° 3, 839-845, 1989.

Baxter J. and J. Atkins, The Fire Came By, Doubleday, Garden City NY, 1976.

Beauford, W., J. Barber and A. R. Barringer, Heavy metal release from plants into the atmosphere, Nature 256, 35-37, 1975.

Beauford, W., J. Barber and A. R. Barringer, Release of particles containing metals from vegetation into the atmosphere, Science 196, 571-573, 1977.

Ben-Menahem, A., Source parameters of the Siberian explosion of June 30, 1908, from analysis and synthesis of seismic signals at four stations, Phys. Earth Planet. Inter. 11, 1-35, 1975.

Bojarkina, A. P., N. V. Vasiljev, G. G. Gluchov, Ju. A. L'vov, V. I. Rezchikov and S. I. Shcheludko, K otsenke kosmogonnogo pritoka tjazhelych metallov na poverchnost' zemli, in Kosmicheskoje veshchestvo i zemlja, 203-206, Nauka, Novosibirsk, 1986.

Boutron, C., Respective influence of global pollution and volcanic eruptions on the past variations of the trace metals content of Antarctic Snows since 1880's, J. Geophys. Res. 85, n° C12, 7426-7432, 1980.

Bradley, J. P., S. A. Sandford and R. M. Walker, Interplanetary dust particles, in Meteorites and the early solar system, 861-895, The University of Arizona Press, Tucson, 1988.

Cecchini, S., M. Galli, G. Longo and R. Serra, Identikit di un'esplosione, Sapere 944, 9-15, 1992.

Chyba, C. F., P. J. Thomas and K. J. Zahnle, The 1908 Tunguska explosion: atmospheric disruption of a stony asteroid, Nature 361, 40-44, 1993.

Cowan, C., C. R. Atluri and W. F. Libby, Possible anti-matter content of the Tunguska meteor of 1908, Nature, 206, 861-865, 1965.

D'Alessio, S. J. D. and A. A. Harms, The nuclear and aerial dynamics of the Tunguska event, Planet. Space Sci. 37, n° 3, 329-340, 1989.

Emeljanov, Ju. M., Radiofotograficheskoje issledovanie srezov derevjev iz rajona padenija Tungusskogo meteorita, in Problema Tungusskogo meteorita, 153-158, Izdatelstvo Tomskogo Universiteta, Tomsk, 1963.

Fast, V. G., Statisticheskij analiz parametrov Tungusskogo vyvala, in Problema Tungusskogo meteorita, part 2, 40-61, Izdatelstvo Tomskogo Universiteta, Tomsk, 1967.

Florenskij, K. P., A. V. Ivanov, N. P. Iljin, M. N. Petrikova and L. E. Loseva, Khimicheskij sostav kosmicheskikh sharikov iz rajona Tungusskoj katastrofy i nekotorye voprosy differentsjatsii veshchestva kosmicheskikh tel, Geokhimija 10, 1163-1173, 1968.

Galli, M., G. Longo, R. Serra and S. Cecchini, La spedizione al luogo della "catastrofe di Tunguska" del 1908, in Dentro la Fisica, 237-252, Clueb, Bologna, 1994.

Golenetskij, S. P., V. V. Stepanok and E. M. Kolesnikov, Priznaki kosmokhimicheskoj anomalii v rajone Tungusskoj katastrofy 1908 g., Geokhimija 11, 1635-1645, 1977a.

Golenetskij, S. P., V. V. Stepanok, E. M. Kolesnikov and D. A. Murashov, K voprosu o khimicheskom sostave i prirode Tungusskogo kosmicheskogo tela, Astronomicheskij Vestnik 11, n. 3, 126-136, 1977b.

Grigoryan, S. S., K voprosu o prirode Tungusskogo meteorita, Doklady Akad. Nauk SSSR 231, n° 1, 57-60, 1976.

Grigoryan, S. S., Motion and destruction of meteorites in planetary atmospheres, Cosmic. Res. 17, 724-740, 1979.

Hills, J. G. and M. P. Goda, The fragmentation of small asteroids in the atmosphere, Astron. J. 105, 1114-1144, 1993.

Hoyt, D.V. and C. Frölich, Atmospheric transmission at Davos, Switzerland 1909-1979, Climatic Change 5, 61-71, 1983.

Hunt, J. N., R. Palmer and W. Penny, Atmospheric waves caused by large explosions, Philos. Trans. Roy. Soc. Lond., A252, 275-315, 1960.

Iljina, L. P., L. M. Slivina, D. V. Demin, V. K. Zhuravlev, L. P. Potekhina, M. N. Levchenko, B. I. Vronskij, O. A. Egorshin and G. M. Ivanova, Rezultaty spektral'nogo analiza prob pochvy iz rajona padenija Tungusskogo meteorita, in Sovremennoje sostojanie problemy Tungusskogo meteorita, 25-27, Izdatelstvo Tomskogo Universiteta, Tomsk, 1971.

Jackson A. A. IV and M. P. Ryan, Was the Tungus event due to a black hole? Nature 245, 88-89, 1973.

Jéhanno, C., D. Boclet, J. Danon, E. Robin and R. Rocchia, Etudes analytique de sphérules provenant du site de l'explosion de la Toungouska, C. R. Acad. Sci. Paris 308, Serie II, 1589-1595, 1989.

Kazantsev, A. P., Vzryv, Vokrug sveta 1, 39-46, 1946.

Kirichenko, L. V. and M. P. Grechushkina, O radioaktivnosti pochvy i rastenij v rajone padenija Tungusskogo meteorita, in Problema Tungusskogo meteorita, 139-152, Izdatelstvo Tomskogo Universiteta, Tomsk, 1963.

Kolesnikov, E. M., A. L. Ljul and G. M. Ivanova, Priznaki kosmohimicheskoj anomalii v rajone Tungusskoj katastrofy 1908 g., II, Astronomicheskij Vestnik 11, n. 4, 209-218, 1977.

Kolesnikov, E. M. and G. I. Shestakov, Izotopnyj sostav svintsa iz torfov rajona Tungusskogo vzryva 1908 g., Geokhimia 8, 1202-1211, 1979.

Kovalevskij, A. L., I. V. Reznikov, N. G. Snopov, A. B. Osharov and V. K. Zhuravlev, Nekotoryje dannyje o raspredelenii khimicheskikh elementov v pochvakh i rastenijakh v rajone padenija Tungusskogo meteorita, in Problema Tungusskogo meteorita, 125-133, Izdatelstvo Tomskogo Universiteta, Tomsk, 1963.

Kresák, L.', The Tunguska object: a fragment of comet Encke? Bull. Astron. Inst, Czechosl. 29, 129-134, 1978.

Krinov E. L., Giant Meteorites, 125-265, Pergamon Press, Oxford ,1966.

Kulik, L. A., Dannyje po Tungusskomu meteoritu k 1939 godu, Doklady Akad. Nauk SSSR 22, n° 8, 520-524, 1939.

Lamb, H. H., Weather, climate and human affairs, Routledge, London, 1988.

Levin, B. Yu. and V. A. Bronshten, The Tunguska event and the meteors with terminal flares, Meteoritics 21, n° 2, 199-215, 1986.

Nazarov, M. A., M. I. Korina, L. D. Barsukova, E. M. Kolesnikov, I. V. Suponev and G. M. Kolesov, Veshchestvennye sledy Tungusskogo bolida, Geochimia 5, 627-639, 1990.

Nesvetajlo, V. D. and N. N. Kovaliukh, Dinamika kontsentratsii radiougleroda v godichnykh kol'tsakh derevjev iz tsentra Tungusskoj katastrofy, in Meteoritnyje i meteornyje issledovanija, 141-151, Nauka, Novosibirsk, 1983.

Passey, Q. R. and H. J. Melosh, Effects of atmospheric breakup on crater field formation, Icarus 42, 211-233, 1980.

Petrov, G. I. and V. P. Stulov, Motion of large bodies in the atmosphere of planets, Cosmic. Res. 13, 525-531, 1975.

Sandford, S. A., The collection and analysis of extraterrestrial dust particles, Fund. Cosmic Phys. 12, 1-73, 1987.

Sekanina, Z., The Tunguska event: no cometary signature in evidence, Astronomical Journal 88, n° 9, 1382-1414, 1983.

Serra, R., S. Cecchini, M. Galli and G. Longo, Experimental hints on the fragmentation of the Tunguska Cosmic Body, Planet. Space Sci., to be published, 1994.

Simkin, T., L. Siebert, L. McClelland, D. Bridge, C. Newhall and J. H. Latter, Volcanoes of the world, Houtchinson Ross Publishing Company, Stroudsburg, Pennsylvania, 1981.

Stuiver, M. and B. Becker, High-precision decadal calibration of the radiocarbon time scale, AD 1950-6000 BC, Radiocarbon 35, n°1, 35-65, 1993.

Tsymbal M. N. and V. E. Shnitke, Gazovozdushnaja modelj vzryva Tungusskoj komety, in Kosmicheskoje veshchestvo i zemlja, 98-117, Nauka, Novosibirsk, 1986.

Turco, R. P., O. B. Toon, C. Park, R. C. Whitten, J. B. Pollack and P. Noerdlinger, An analysis of the physical, chemical, optical and historical impacts of the 1908 Tunguska meteor fall, Icarus 50, 1-52, 1982.

Valdré, G. and K. Korlevic, New method for the study of airborne particles based on tree-resin trapping and analytical electron microscopy, Ultramicroscopy 49, 382-386, 1993.

Vasiljev, N. V, Ju. A. Lvov, B. I. Vronskij, Ju. A. Grishin, G. M. Ivanova, T. A. Menjavtseva, S. N. Grjaznova and P. P. Vaulin, Poiski melkodispersionnogo veshchestva v torfah rajona padenija Tungusskogo meteorita, in Meteoritika 32, 141-146, 1973.

Vasiljev, N. V., Istorija izuchenija problemy Tungusskogo meteorita v poslevojennyje gody (1958-1969), in Meteoritnyje issledovanija v Sibiri, 3-22, Nauka, Novosibirsk, 1984.

Vasiljev, N. V., Istorija izuchenija problemy Tungusskogo meteorita v poslevojennyje gody (1970-1980), in Kosmicheskoje veshchestvo i zemlja, 3-34, Nauka, Novosibirsk, 1986.

Vasiljev, N. V., Istorija izuchenija problemy Tungusskogo meteorita v poslevojennyje gody (1980-1985), in Aktual'nyje voprosy meteoritiki v Sibiri, 3-31, Nauka, Novosibirsk, 1988.

Vasiljev, N. V., G. F. Plekhanov and G. V. Andreev, O mezhdunarodnoj programme issledovanija Tungusskoj katastrofy 1908 goda, in Neperiodicheskije bystroprotekajushije javlenija v okruzhajushej srede, 201-207, Tomsk, 1990.

Whipple F. J. W., The great Siberian meteor and the waves, seismic and aerial, which it produced, Quart. J. Roy. Meteorol. Soc. 56, 287-304, 1930.

Zigel', F. Ju, K voprosu o prirode Tungusskogo tela, in Meteoritnyje i meteornyje issledovanija, 151-161, Nauka, Novosibirsk, 1983.

Tables

Table 1.
Trees and resin samples examined.

A) Resin samples from tunguska branches

N°	Tree	Years	branches	resin samples
1	Siberian spruce (Picea obovata)	160	3	19
3	Siberian spruce (Picea obovata)	130	2	5
4	Siberian spruce (Picea obovata)	190	3	10
5	Siberian pine (Pinus cembra)	110	1	7
6	Siberian spruce (Picea obovata)	110	2	6
7	Siberian spruce (Picea obovata)	110	2	7
8	Siberian spruce (Picea obovata)	105	1	5
		TOTAL	14	59

B) Resin samples from control trees

N°	Tree	branches	resin samples
10	Tomsk Siberian spruce (Picea obovata)	6	8
2	Roots from Siberian larch (Larix sibirica)	3	4

Table 2.
The 28 elements present in the particles from Tunguska tree-resin (more than 1 % in at least 1 particle). The equipment used cannot reliably detect elements with Z < 10 or quantities < 1 %.
Low-Z elements       Na Mg Al Si P S Cl K Ca Fe
High-Z elements       Ti Cr Mn Co Ni Cu Zn Br Sr Ag
                               Cd Sn Sb Ba W Au Pb Bi

Table 3.
Particles in the resin samples from Tunguska branches. Average dimensions of the particles: 4.5 µm x 3.1 µm (0.5 µm x 0.5 µm < 66 % <= 3 µm x 3 µm; 3 µm x 3 µm < 29 % <= 10 µm x 10 µm).

Period (years)	Number of samples	Resin surface (cm²)	particles			density (particles/cm²)
Period (years)	Number of samples	Resin surface (cm²)	High-Z	Low-Z	Total	High-Z	Low-Z	Total
1885-1901	17	4.46	20	1161	1181	4	260	265
1902-1914	25	7.60	463	2893	3356	61	381	442
1915-1930	17	5.40	35	1282	1317	6	237	244
TOTAL	59	17.46	518	5336	5854	30	306	335

Table 4.
Elements related to the Tunguska event

Войти в систему

G. Longo, R. Serra, S. Cecchini and M. Galli, SEARCH FOR MICROREMNANTS OF THE TUNGUSKA COSMIC BODY Planetary and Space Science 42, n.2, p. 163-177; 1994

G. Longo, R. Serra, S. Cecchini and M. Galli, SEARCH FOR MICROREMNANTS OF THE TUNGUSKA COSMIC BODY
Planetary and Space Science 42, n.2, p. 163-177; 1994