PADVARNINKAI, THE EX-SHERGOTTITE…, appeared in Sphaerula 2, 25 (1998-2001) and did not draw citations. So I put it on Internet.
The second shergottite (yes, it had been so) became almost forgotten because Lithuania was incorporated into the USSR for half a century. The main mass was first in Kaunas and then in Vilnius, but scientists outside the USSR could not visit the meteorite and of course could not collaborate with the Lithuanian colleagues, but it was rather difficult for Russians too. So the meteorite could and so did not get the proper attention.
Padvarninkai with its triple lithology suggests an involved creation process, which was surely long too. But this paper was not interested in mineralogy; we analysed the elemental composition. That was peculiar too.
Here it comes.
PADVARNINKAI, THE EX-SHERGOTTITE (A STUDY ON BULK COMPOSITIONS)
Ágnes Holba & B. Lukács
CRIP RMKI, H-1525 Bp. 114. Pf. 49., Budapest, Hungary
ABSTRACT
Between 1929 and 1967 the Padvarnninkai meteorite was regarded as shergottite according to the definition of Tschermak, having (clinopyroxenic) basalt together with maskelynite; it was the second member of the group. In 1967 it was reclassified as howardite, so it was not considered when Martian origin of the shergottites became seriously suggested. Now the majority opinion is that it is a polymict eucrite. In this paper we compare the Padvarninkai bulk composition (according to the original measurement of Kaveckis) to that of 21 diogenites, 21 howardites, 79 eucrites (of which 18 is polymict), 16 ureilites, 7 basaltic lunars and 14 shergottites. The result is that the Padvarninkai point does not fit into any of these groups. Our guess is that the special prehistory is responsible.
1. INTRODUCTION
The story of the reception of the Padvarninkai meteorite was as complicated and involved as that of the place of deposit of its main mass; and as the prehistory of the meteorite. Polymict meteorites are not rare; but generally polymict means dyomict; one main mass contains clasts from another (and brecciated ones are brecciated from two). However Padvarninkai contains three different lithologies [1]. (Note: trial collisions are generally much rarer than dual ones.) And, in spite of being classified as a polymict (really, then, triamict) eucrite, according to old measurements, some part of it does not looks like an eucrite at all. This paper is devoted to Padvarninkai, and especially to the contrast between its bulk composition and that of eucrites and shergottites.
In recent years cca. a dozen meteorites have been accepted as Martian. The degrees of evidence are various, which is natural as far as wide-scale sampling is not done. Lunar meteorites were identified after 1969, when Apollo, and later Luna, probes brought a lot of martian rocks. Up to now we do not have Martian rocks at hand, except perhaps some meteorites, which we just would like to check. Almost quarter century old Viking and present Pathfinder measurements do help, but the areas investigated are very small, the rocks are only a few, and of course the information gathered is moderate compared to direct laboratory measurements.
What is really interesting is the relatively large number of Martians. Let us compare the numbers of Martian to Lunar meteorites. The canonical number of Martians is 12, and as a relatively new and very detailed review see Ref. [2]. Now, we can make Table 1 about some characteristics we now feel important from the viewpoint of the present paper:
Name |
Time1 |
Weight,g |
Arrival2 |
Type3 |
Glass4 |
|
|
|
|
|
|
The “Canonic. 12" |
|
|
|
|
|
ALH84001 |
1984 |
1,931 |
Not |
I |
YM |
ALHA77005 |
1977 |
482 |
Not |
IS |
YM |
Chassigny |
1815 |
c.4,000 |
Seen |
I |
Y |
EETA79001 |
1979 |
7,942 |
Not |
IS |
YM |
Gov. Valadares |
1958 |
158 |
Not |
IC |
Y |
Lafayette |
1931 |
800 |
Not |
IC |
Y |
LEW88156 |
1988 |
13 |
Not |
IS |
YM |
Nakhla |
1911 |
9,900 |
Seen |
IC |
Y |
QUE94201 |
1994 |
12 |
Not |
IS |
YM |
Shergotty |
1865 |
4,983 |
Seen |
IS |
YM |
Y793605 |
1979 |
16 |
Not |
IS |
YM |
Zagami |
1962 |
c.18,120 |
Seen |
IS |
YM |
______ |
__ |
___ |
_ |
_ |
|
Newly Suggested Martians |
|
|
|
|
|
Dar al Gani5 |
1997 |
c.6,300 |
Not |
IS |
Y |
Dhofar019 |
2000 |
1,056 |
Not |
IS |
YM |
Los Angeles6 |
1999 |
698 |
Not |
IS |
YM |
NWAfrica4807 |
<2001 |
28 |
Not |
IS |
YM |
SayhalUhaymir8 |
1999 |
10,560 |
Not |
IS |
YM |
______ |
__ |
___ |
_ |
_ |
|
Padvarninkai |
1929 |
3,858 |
Seen |
I |
YM |
|
|
|
|
|
|
_______________
1 Fall, find or description time
2 Arrival seen or not
3 I: Igneous rock, IS: shergottite, IC: clino-pyroxene; O: others
4 Glass: Y: observed, YM: maskelynite is recorded; N: none
5 At least 4 fragments, N° 476, 489, 670 & 735
6 Two fragments, 001 & 002
7 Another, N° 817, 104 g, IC, too recent for details
8 4 fragments, N° 005, 008, 051 & 094
Literature: [2]-[9]
Table 1
12 accepted and several seriously suggested Martian meteorites vs. Padvarninkai.
We can see that the number is rapidly growing from 1997, with finds. So the above Table is not yet complete. However even Table 1 contains 12 "canonical" Martian meteorites, with total mass 48 kg, + >4 suggested with mass 19 kg, of which 4, with 37 kg total mass, arrived in the last 200 years. Now, as the above list is incomplete, maybe we have the right to use one collection, but world's biggest, for lunar meteorites, the Tokyo NIPR Antarctic collection. In 1995 the number of collected meteorites was above 8,000, and that of the classified ones just above 3000 [10]. Now, this 3,000 included 7 lunar meteorites, with 1.2 kg total mass. A recent (2001) world-wide compilation [11] gives 21, with total mass c. 8 kg.
While the surface area of Mars is bigger than that of Moon, the escape velocity is much higher and not all fragments are collected by Earth. So one would expect relatively much more lunar meteorites.
The solution of this inproportionality will find its explanation when manned surveys reach Mars. One possible solution is that some of the above meteorites are not Martian; but even if one believes in this, no serious suggestion exists which ones to leave out. The above dozen forms a tight group; not only homogeneous in several characters and not only impossible to put them into other meteorite groups (although ALHA84001 was first classified as diogenite), but also generally impossible to accept any other imaginable source for them, at least in the Solar System. This question will be detailed in due course.
However these arguments seem to hold, more or less, for Padvarninkai as well. We are not arguing for Padvarninkai as a Martian meteorite, since we feel that the class is too populous even now. However we claim that Padvarninkai is remarkably similar to the recently accepted Martians, and, what really counts, some of its composition data do not fit into the meteorite class whither it is relegated now.
All arguments will be made in due course. But first we note that it was rather a historical accident that Padvarninkai did not even get the chance to start in the competition for the Martian origin; the time was too short between Zagami's fall in 1962, proving finally the extraterrestrial origin of shergottites (and so of Padvarninkai, accepted then as shergottite, the second such on Earth), and the expulsion of Padvarninkai from the tiny shergottite grouplet of 3 in 1967, on the ground that Shergotty and Zagami were nearer to each other than to the protagonist of this paper (which was true but among 3 generally 2 are nearer).
In Sect. 2 we recapitulate why the great majority of Martian meteorites contain maskelynite. Sect. 3 recapitulates the history of opinions about Padvarninkai. In Chap. 4 we tell something about composition maps, a method we apply in this paper. Sect. 5 gives the maps of the biggest components Mg, Fe, Ca and Al (and, implicitly Si, of course), while Sect. 6 and 7 deal with the alkalies. Sect. 8 is summary and outlook. Theories and some technical details have been relegated into the Appendices.
2. ON THE CREATION MECHANISM OF MARTIAN METEORITES
The existence of SNC and related meteorites is a fact; their Martian origin follows from theories. We do not want to go into too much details; however for further argumentation it is necessary to see the great role of shergottite.
A substantial part of the Martians in Table 1 is classified even now as shergottites. The traditional definition of shergottite comes from the Rose-Tschermak-Brezina classification [12]. According to this, shergottite is a texture containing clino-hypersthene and maskelynite. The class was suggested by Tschermak himself and named after the Shergotty meteorite (the generally cited paper is [13], but the analysis was first published by E. Ludwig in Tschermaks Min. Petr. Mitt. 1, 55 (1871); Die Analysis des Meteorites von Shergotty, analyst E. Lumpe), which, for 64 years, was the only shergottite at our reach. McSween & Stolper formulates the relation of shergottite, compared both to eucrites (asteroid basalts) and terrestrial basalts as ([14]; see also [15]): "except for the presence of maskelynite, and unlike the eucrites, the shergottites are virtually identical with tipical terrestrial basaltic rocks..."both [shergottite and terrestrial] contain magnetite, which is evidence that both evolved at higher oxidation states than the eucrites...", and so on. Summarizing, proto-shergottite formed as igneous rock on a planetary body comparable to Earth for gravity, presence of some atmosphere &c., and then underwent a shock turning plagioclase into maskelynite. Obviously from a planetary body comparable to Earth only a big impact can throw out the rock. Note that recently [16], [17] doubts arose about the nature of maskelynite: Chen & El Goresy seem to prefer quenching from dense melt (which is still a product of shock), and they tell that not Shergotty but Chassigny contains diaplectic plagioclase. Argumentation is going; see e.g. [18] for the opposite opinion. While the question is important, obviously will not be decided here; we only call the attention to the ambiguity about source circumstances.
Maskelynite is not characteristic for lunar meteorites. So we can guess that the planetary body producing shergottites must be bigger than Moon. Iovian and Saturnian moons differ a lot for composition. Then in the Solar System only the inner planets remain for source. But Mercury is deep in the gravity well and the big escape velocity + dense atmosphere would prevent the fragment to leave Venus. Then remains Earth and Mars. Because of high escape velocity plus several other considerations discussed in Refs. [14] & [15] Mars is more probable. Observe that all of the "canonical dozen" have glass (and at least in 8 it is maskelynite), so they got serious shock. In addition, compositions are more similar to Viking and Pathfinder data than to terrestrial or HED compositions. So now we do not have anything against the explanation that they were knocked out from the Martian surface.
To be sure, maskelynite in itself is not an evidence for Martian origin. We would not expect it to be, indeed. It is true that a small impact is enough to dislodge a fragment from an asteroid; still from time to time asteroids can suffer big impacts. Only, the majority of other meteorites are expected not to contain maskelynite. And it is so, indeed. Maskelynite appears in a fragment of H, L and LL chondrites from petrologic class 5 upwards [10], and Binns [19] reports 33 chondrites in the British Museum collection with maskelynite. In addition, Mittlefehldt [20] mentions eucrite ALHA81313 containing maskelynite.
So we cannot tell that the presence of maskelynite is the sufficient condition for a meteorite being Martian, since chondrites cannot be Martian. Also, it is not a necessary condition, see the nakhlites and Chassigny in Table 1. Still, there is a strong correlation between maskelynite and Mars, even if we do not yet completely understand the exact nature of this correlation.
3. A VERY BRIEF HISTORY OF PADVARNINKAI AND ITS RECEPTION
Padvarninkai fell just after midnight of 9th February, 1929. It seems tht the original stone fragmented in the atmosphere; 11 pieces were retrieved. The path of the meteorite is excellently observed; details of the fall are reported in Ref. [21] which is written in Lithuanian and difficult to find now because of the complicated XXth century history of Lithuania. Then 2 analyses were made [21], [22], which show the usual differences. Kaveckis' measurement is the more complete one, including Fe°, S and C; we will use those data. The measured elementary compositions were quite similar to that of Shergotty, and photographs clearly show the maskelynite. The conclusion of Ref. [22] is clearly that Padvarninkai is a shergottite meteorite.
With this, the shergottite "class" grew to 2 (two) meteorites (Shergotty and Padvarninkai). This fact, however, did not provoke too much agitation because the origin of shergottites was still unknown. Although the number 2 was too few for real statistics, it was enough to discuss which variations are possible between individual shergottites. It seems, however, that this did not really happen. (It is interesting to note that, while the two analyses of Padvarninkai, of Kaveckis vs. Busz & Biel, fairly coincide for simple molecules, the calculated mineral composition differs more, specially for ferrous pyroxene and (K,Al) and (Ca,Al)-silicates.)
In 1940 the USSR gave back the old Lithuanian university and capitol city Vilnius to Lithuania, incorporating the latter simultaneously. Next year Germany occupied Lithuania, and in 1944, the USSR captured back the country. In demolishing and deportation events in these 4 years the university personnel, libraries and collections were disturbed. In the post-war period Vilnius inherited most roles of Kaunas. So the Padvarninkai meteorite got much less attention than Shergotty.
In 1962 fell the third shergottite meteorite, Zagami. Then, in 1967, Binns [10] investigated the relations among the 3 meteorites, with the result that in some respects Shergotty and Zagami together oppose Padvarninkai. In a few quantitative data Shergotty and Zagami are closer to each other than to Padvarninkai. To be sure, in 1967 maskelynite was known only in these 3 achondrites, while there were some chondritic examples. of which Binns chose Chateau-Renard. Then he found that Shergotty & Zagami contain two monocline pyroxenes, namely pigeonite and augite. In contrast, Padvarninkai contains two monocline pyroxenes as well, but they are ferrohypersthene and subcalcic ferroaugite. Comparing the refractive indices the sequence was Padvarninkai>Shergotty~Zagami>Chateau-Renard. Plus Padvarninkai is brecciated (Chateau-Renard, if not brecciated, is veined [10]).
Hence Binns concludes that "maskelynite occurs in two distinct [our emphasis; Á.H. & B.L.] kinds achondrite". And that "Padvarninkai is probably best regarded as a brecciated howardite [our emphasis; Á.H. & B.L.], but Shergotty and Zagami together constitute a unique class"; of course then that class is shergottite.
Half of the proposition was accepted immediately, and Padvarninkai never was considered shergottite. The other half was not successful. Of course, Padvarninkai cannot be regarded as a howardite, brecciated or not, we will go to this point in due course. Instead now the consensus is that it is a polymict eucrite.
In the remaining part of this paper we show that, at least for bulk composition data, Padvarninkai does not seem an eucrite either. For texture we will use only the 2 generally accepted facts that i) it contains maskelynite, and ii) it is brecciated or polymict.
Henceforth we handle Padvarninkai as a disjoint individual and compare it with all the groups which can be suspected in any sense. We consider our work as a generalisation of Binns, on a much greater database.
4. PADVARNINKAI AMONG "BASALTIC" ACHONDRITES: THE METHOD
For general compositional comparisons we have formed a database containing 968 meteorite data (a few individuals dually or even triply represented). It contains 21 independent constituents which are listed in Appendix B. Only such meteorites were taken whose available datasets contain all constituents or the lacking ones can be reconstructed somehow. The data have been transformed to the same forms but no effort was made to remove systematic errors; that will be the topic of subsequent papers. The 4 catalogs yielding the overwhelming majority of the raw data are considered reliable. (We, however, have detected systematic differences.)
Our database is composed as follows.
1) The NIPR Catalog [10]; 525 Antarctic and 24 non-Antarctic meteorites, an extremely homogeneous database because all but 13 meteorite was measured in the NIPR laboratory by H. Haramura. Irons we have ignored.
2) Jarosewich's "catalogue" [23], some 230 meteorite (we again have ignored irons + a few inclusions).
3) The compilation of Urey & Craig [12]. The source of that compilation was not homogeneous at all; but Urey and Craig cross-checked the data in various ways and refused some 2/3 of the data available for them. One may hope that the remaining part is not too bad.
4) Wiik's small, but extremely homogeneous dataset of 31 measurements [24] (we had to leave out 1, not having FeO content).
5) Some two and half dozen other meteorites collected by us from the literature. The now relevant part of the list will be given in Appendix B.
It was possible to estimate the overall mean deviations of the datasets, at least in the sense that error propagation can be calculated in statistical sense. That point is again relegated to Appendix, but we note that the statistical errors of the 4 "catalogs" are conform to expectation: [10]<[23]<[24]<[12].
In this paper we use the following groups for comparison:
1) Padvarninkai.
2) Shergottites: the old class of Padvarninkai.
3) Polymict eucrites: the new group of Padvarninkai.
4) Other eucrites.
5) Howardites: Binns' suggestion for Padvarninkai.
6) Diogenites, as the remaining HED's.
7) Lunars, as analogons of eucrites [14], [15].
8) Ureilites, as relatives (in some sense [25]) to HED.
9) "Aberrant" HED's, explained below.
The "aberrant" class contains 4 HED's, Massing (How) and Pavlovka, Serra de Mage and Zmenj (Euc), all from [12], who behave on the composition maps as not proper HED's (as we shall see). We note that Serra de Mage is in the [10] and [23] catalogs too, there without this anomaly. Any explanation is welcome; now we continue.
The maps of Sects. 5-7 will demonstrate that there is, indeed, a strong correlation between petrology and chemical composition. While this is not too surprising, the true reasons are matter of theory. (Of course, if there was a unique parent body of eucrites, one expects some common chemical characteristics too.) We do not want to use too much theory. Take the maps simply as statistical facts, and conclusions as probability statements.
5. Fe, Mg, Ca AND Al: DIFFERENCES BETWEEN HED'S
Henceforth data are always in weights, normalised to Si. Fig. 1 is total Fe vs. Mg. For comparison to chondrites see Refs. [26]-[33].
The different groups of achondrites more or less separate on this map. Compared to chondrites ureilites are hyper-magnesians. Diogenites are almost at chondritic Mg level; for total Fe they are below, but this is because of the lack of metallic Fe. For FeO vs. Mg they would almost coincide with the chondrites, although they seem to be slightly impoverished in Mg. Note that this group is the "chondritic type achondrites" [12] in the old the Rose-Tschermak-Brezina classification. Indeed, for composition the only big difference is the lack of FeNi metal.
Although we do not want to depend on theory too much here, theoretical explanation is relatively easy in this first point. In a parent body big enough to melt metallic Fe goes into the core and chondrules vanish. If differentiation stops here, we get diogenite-type achondrites [27], [28], [30], [34], [35].
Of the other HED's howardites are more magnesium-free, and eucrites, polymict or not, are the poorest in Mg, together with lunars [26], [29]. Our guess is that late volcanism is more and more difficult, so heavy and highly solidifying Mg/Fe silicates erupt less, while they are replaced more and more by Ca/Al/Na-silicates [36], [37]; we shall see at Fig. 2. But theory, or not, the trend can be seen, although a few eucrites mix with howardites. Polymict eucrites cannot be told apart from other eucrites on this map.
It is worthwhile to note here that now it is a general belief that howardites are not a distinct class per se but breccias of diogenites and eucrites, and their intermediate compositional data come from physical mixing. The idea goes back to [38], and indeed there a number of comparisons are made which all show an approximate linear relationship in the D-H-E sequence. On the other hand, progressing differentiation would produce the same approximate linearity, at least for major elements, in the sequence D->H->E (Al, Ca and Na are substituting Mg and Fe because of lightness and low melting point, Ti is increasing for less viscosity &c.). And note that K. Yanai reported an unbrecciated, holocrystalline meteorite with good "howardite" composition [39]. (The exact meaning of "howardite composition" is seen on the maps; howardites seem to form a nucleus somewhere between diogenites and eucrites, and not simply filling the gap linearly, although there are interpolating ones too.) If so, then "howardites" may cover 2 groups of different origins; however let us pass.
Shergottites cover a great area, but the majority contains more Mg than diogenites do.
Padvarninkai is at the edge of the eucrite group, but also at the edge of howardites and of lunars; but there is also a shergottite very near. This check does not discriminate among eucrite, howardite or shergottite origins of Padvarninkai.
Fig. 2 is Ca vs. Al. Both components make the silicates light and of low freezing point. Now, all HED's + ureilites occupy a fan-shaped area, with the original chondritic ratio [29]. Diogenites and ureilites are near to the origin, while howardites are in middle positions, and eucrites (polymict or not) at the top. It seems as if we saw the gradual substitution of Fe and Mg with Al and Ca. But again, the facts are independent of theory.
While HED's keep the original Ca/Al ratio, shergottites and, in lesser extent, lunars deviate from the original ratio. For terrestrial basalts also deviation is seen. Since amongst the planetary bodies whose basalts we have the mass sequence is Earth>Mars>Moon>Vesta (?), one can imagine that some differentiating mechanisms are more and more effective with increasing mass. Here we do not want to guess the proper mechanism.
Now observe that Padvarninkai also deviates from the proportionality together with shergottites and lunars, some shergottites not too far. On this map Padvarninkai does not seem either to eucrite or to howardite.
Figs. 3 & 4, total Fe vs. Al and Ca, are other projections of the 4 dimensional map. Polymict eucrites cannot be distinguished from other eucrites; Padvarninkai is at or near to the edges of eucrites, howardites and lunars, but never inside.
In the next Chapter we will look at further maps. However observe even now that Padvarninkai cannot be confused with diogenites or ureilites, so we will generally ignore these two groups in the comments. Note also that while howardites marginally overlap with a few eucrites, the groups themselves can be distinguished.
6. POTASSIUM MAPS
Chondrites are poor in potassium. It is more abundant in HED's but with serious mean deviation. Now Fig. 5 shows K vs. Mg. The 3 HED groups are distinct, polymict eucrites again cannot be told apart from other eucrites. Higher K content is rare among diogenites.
On this map the highest K content is in an "aberrant HED", Massing. The next is Padvarninkai, then comes Shergotty. The next two, near each other, are the "aberrant" Pavlovka and Zmenj; the "aberrant" Serra de Mage does not yet seem aberrant on this map. Fig. 6, the (K,Al) map, is similar. Padvarninkai is not near to any group, while it is halfway between Shergotty and the "aberrant howardite" Massing.
This Chapter did not settle anything except that the Padvarninkai point [12], i.e. the measurement of Kaveckis, is nowhere near to any eucrite, and is halfway between two XIXth century measurement for Shergotty and Massing.
7. SODIUM; A VOLATILE MAJOR CONSTITUENT
Fig. 7 is the Na vs. Mg plot. We already have discussed the Mg levels. Polymict eucrites do not separate from other eucrites even on this plot. The highest Na values are shown by the 4 "aberrant" HED's, and if we do not want to try to explain it, it is better to pass. (This was the reason to create this "aberrant" group.) Shergottites cover a wide range, but they fit to a strip which is either a hyperbole, or a decreasing slope (statistics is not large enough to discriminate). Although we would like to restrict ourselves from theory, this is not strange at all. Mg-silicates are heavy and freeze at high temperatures, Na-silicates just the opposite. Martian basalts must be as wide a class than terrestrial ones [29]. On Earth we have lherzolitic and komatiitic basalts with a lot of Mg and negligible amount of Na, and, say, andesites just the opposite.
However now Padvarninkai fits also to this strip. On this map Shergotty is hardly distinguishable from Padvarninkai; but EETA 79001 is quite near too.
Fig. 8 is Na vs. Al. For structure the Figure is similar to Fig. 2, although not so simple. Diogenites are near the origin, howardites are upper and to the right, eucrites (any) in the upper right corner, but on this map, exceptionally, the polymict eucrites somewhat tend to the left side of the other eucrites. Still, the overlap is too large for discrimination. Aberrant HED's are everywhere at the top. Shergottites and Padvarninkai again go left from the main field, but now lunars do not.
Fig. 9 is Na vs. total Fe. HED's cannot be distinguished from each other, but shergottites and lunars are right from the main field (more iron). Padvarninkai merges among the shergottites.
And Fig. 10 is the two alkalies, K vs. Na. Almost all meteorites group in the neighbourhood of the origin; 3 aberrant HED's and some shergottites go to the right of the main field. In the upper right corner there are only 3 meteorites: Padvarninkai in equal distances from, although not quite between, the aberrant Massing howardite and the Shergotty shergottite.
Now, briefly, back to theory. Sodium is a volatile, at least in HED context. According to present consensus, HED's are asteroid basalts. The biggest asteroids are spheroids, so were molten for a while, and 4 Vesta is covered with basalt. So let us say that HED's originated on Vesta, or on a destroyed asteroid roughly similar to Vesta.
Now, in the time of melting T was cca. 1500 K. As for escape velocity
vesc = (2GM/R)1/2 (7.1)
and for a sphere of 3 g/cm3 density and roughly Ceres radius (the biggest surviving asteroid) vesc is cca. 0.65 km/s, even smaller for Vesta. On the other hand, for ther average thermal velocity
vth = (3kT/m) (7.2)
While for a silicate molecule vth is not too large, obviously a Vesta-type body cannot retain Na2O, for which vth is cca. 0.8 km/s at melting temperature. So, even without detailed calculation for silicate decomposition, one expects serious Na loss for the molten stage of asteroids. Meteorite statistics are quite unequivocal about this. In Ref. [40] we evaluated the NIPR data about Na/Si values. We know the "galactic", or "cosmic" abundance [41], cca. 0.046. "Early" (low PC) chondrites are cca. the 3/4 of this, either Na2O being volatile, or forming first the mixed oxide NaHO ş NaOH which is even more volatile (molecular weight being 62 for Na2O but only 40 for NaHO). C chondrites have even lower Na abundance, maybe via the competition of H and Na. For this point see also [42].
Now, because Na-silicates are light (and of low melting point) one would expext enrichment in mature volcanism, just as it is on Earth, where in the upper crust now Na is more abundant than Mg while for cosmic abundance Na/Mg is below 0.1.
However it is oppositely for meteorites. NIPR data give just below 1 weight % of Na2O for chondrites, while below 0.1 % in average for diogenites. In the old Rose-Tschermak-Brezina classification our diogenites (e.g. Johnstown and Tataouine) classified as "chondrite type achondrites", and the biggest difference for composition was the less Fe. Now another difference between diogenites and chondrites is the low Na level. If diogenites are "early lavas", e.g. analogons of komatiites, then the Na loss from hot stage explains the impoverishment.
However volcanism do enrich the lavas (and so the rejuvenated surface) in light elements; compared to the almost Na-free first stage. In howardites the Na level is double or treble than in diogenites [10]; and in eucrites it almost climbs up to the chondritic value; it is 0.3-0.5 %. Similar for lunars, whence we can get only the fragments of the very last volcanism.
Now, Mars is a major planet, with vesc » 5 km/s. This being 6 times the thermal average velocity of Na2O of the hot stage, according to planetologic thumb rules Mars could keep all his sodium. And the roughly terrestrial Na level is seen both in shergottites and in Viking and Pathfinder data.
This point of argumentation is almost theory-free, depending on fundamental physics. But also, it is supported by the great amount of very reliable and homogeneous NIPR standard wet measurements. HED meteorites (and ureilites) are very poor in sodium, even eucrites and lunars far below terrestrial level.
|
Fig. 1: Fe/Si vs. Mg/Si; explanation in text. P. is among eucrites.
|
Fig. 2: As above, but Ca/Si vs. Al/Si. P. is rather lunar.
|
Fig. 3: As above, but Fe/Si vs. Al/Si. P. is between eucrites and lunars.
|
Fig. 4: Fe/Si vs. Ca/Si. P. is again bw. eucrites and lunars.
|
Fig. 5: K/Si vs. Mg/Si. P. is far from the eucrite cluster.
|
Fig. 6: K/Si vs. Al/Si. P. is far from any group.
|
Fig. 7: Na/Si vs. Mg/Si. P. is among shergottites.
|
Fig. 8: Na/Si vs. Al/Si. P. is bw. eucrites and shergottites.
|
Fig. 9: Na/Si vs. Fe/Si. P. is far from any group.
|
Fig. 10: K/Si vs.Na/Si. What is P.? (But quite alkalic.)
8. OUTLOOK: WHAT MAY BE PADVARNINKAI (IF ANY)?
Binns was completely right: Padvarninkai is not a regular shergottite from compositional viewpoint either. However it (either the piece measured by Kaveckis [21], [22], or that by Busz & Biel [22]) is not a regular HED, on the other hand.
In Binns' analysis, the pyroxenes of Padvarninkai were analogous to those of Shergotty and Zagami, but quite different for compositional details. Now we see the same for bulk composition. However what we do see is disturbing enough.
In the 4 dimensional subspace (Mg,Fe,Al,Ca) the Padvarninkai point is always at the edge or out of any of the HED groups. With some luck or assumed anomaly Padvarninkai could be a eucrite (although the Padvarninkai point is nearer to some non-polymict than to polymict ones), but it very probably is not a howardite on pure statistical grounds. (Of course, this statement is empty if howardites do not exist on their own rights, but only as mixtures of eucrites and "something else", but cf. Ref. [39]; and if they are mixtures of eucritic and diogenitic material, they must be between the eucrite and diogenite spots, where the Padvarninkai point is definitely not located on the maps of this paper.) True, the point does not seem to be at a characteristic shergottite site either. The (Al,Ca) map suggests, however, that the Padvarninkai composition is more differentiated than any asteroid basalt (HED).
And with the alkalines the picture changes. For Na/K maps Padvarninkai is far from all HED's of the NIPR and Jarosewich databases, but it is similar to some shergottites (including Shergotty), and to 4 "aberrant" HED's of the Urey-Craig compilation.
There is a possibility that the four "aberrant" HED points are measurement errors, and so is Padvarninkai. While this is possible, such explanation seems rather desperate. Namely, it is true that we took Padvarninkai from the Urey-Craig compilation, and also the four aberrant HED's and Shergotty are there. But
1) Urey and Craig did not measure these points. The 6 crucial meteorites were measured at 6 localities (Munich (?), Kaunas, Berlin, Rio de Janeiro, Vienna and somewhere Russia) by 6 different analists between 1871 and 1929.
2) We, as well as Urey and Craig, used Kaveckis' measurement for Padvarninkai. Now, the alternative measurement [22] gives even higher value for the alkalis, so it would be even more contradictory to eucritic or howarditic affiliation.
3) True, the aberrant Serra de Mage Na2O value seems to be ruled out; its "terrestrial" 1.59 % contradicts to Jarosewich's 0.30 % [23] or to the Yanai & Kojima value 0.19 % [10]. (We kept all 3 points for Serra de Mage.) However observe that the "aberrant" Serra de Mage was not similar to Padvarninkai otherwise than a high Na value. The only aberrant HED showing a tendency "to be similar" on the maps to Padvarninkai was the howardite Massing, for which we do not know any specific counterargument (only that the point is highly improper and improbable for a howardite).
Of course, new measurements would be welcome (also on Massing). If the Kaveckis and/or Busz-Biel measurements represent Padvarninkai even approximately, then it is rather hard to imagine its parent body in the Solar System, and this is surely true from phenomenologic & statistical viewpoint.
Of course, Padvarninkai is brecciated (or polymict). This may help, because then different parts may have different origins; and there are even 3 lithologies [1]. However surely, Kaveckis chose one piece, have been it any. But this one piece points to an asteroid parent body from one aspect and to a major planet from another.
And note that in some sense the opinion that Padvarninkai is not a "usual" polymict eucrite is haunting the literature.
Summarizing (for some details see App. A): it seems as if as many as 4 impacts may have happened on the components of the thing we call Padvarninkai; the last one produced the rock breaking up into 11 pieces just after midnight (Central European Time) on February 9, 1929 in the atmosphere of Lithuania. The complicated prehistory may account for some of the anomalies. However standard wet analyses are generally local, using small pieces, and even then the bulk composition of the measured spot of Padvarninkai seems to remain unique. The Na and K values are rather high for diogenitic or eucritic matter or even for a serious contribution of such in the measurement.
ACKNOWLEDGEMENTS
Earlier discussions with Drs. K. Yanai and H. Kojima about the NIPR meteorite collection, Sz. Bérczi about petrology & mineralogy and K. Martinás about thermodynamics are thanked. Partly supported by OTKA grant T/026660.
APPENDIX A: THEORIES AND SUGGESTIONS ABOUT THE ORIGIN OF PADVARNINKAI
Since for 38 years Padvarninkai was considered a sister of Shergotty, the prehistory of Padvarninkai theories was the opinion on Shergotty. We can start with Tschermak.
Tschermak, first substantial investigator of Shergotty, declared it volcanic [43], being clearly basaltic. Some other meteorites, as e.g. the Moravian Stannern, were also basaltic, but Shergotty was unique for having maskelynite.
Clinopyroxenite eucrites and the non-maskelynitic part of Shergotty were indeed so similar to terrestrial basalts that Astronomer Royal of Ireland, R. Ball, claimed them to be terrestrial [44]. He carefully discriminated between shooting stars and meteorites. Shooting stars are connected to comets, while meteorites are not. In Ref. [44], already after Shergotty's investigation (the book was first published in 1885), he concluded that meteorites are mainly rock, moreover, volcanic rock, closely related to terrestrial ones. Since planetary surfaces are not too similar to terrestrial ones, probably meteoritic basalt originated from Earth in earlier times of greater volcanic activity which meet now from time to time Earth's orbit.ef. [44]. (Shergotty was indeed quite terrestrial-like; see also [14].) However terrestrial escape velocity is too large for volcanic eruptions, so Padvarninkai was not considered seriously terrestrial in any time.
In 1967 Binns reclassified Padvarninkai as brecciated howardite. In subsequent years the idea propagated that HED's would be fragments of substantial asteroids, most probably of 4 Vesta. Indeed, Vesta shows mainly eucritic reflectance spectrum, and one diogenitic spot [45]. And Vesta, while fairly spheroidal, shows an asymmetry simplest to interpret as a big crater of a splintering collision. Maybe that impact dug out the diogenic lower levels. (However howarditic surface is not seen.) Later Padvarninkai has been reclassified as polymict eucrite.
In the 90's some phenomenologic information has accumulated, without complete explanation of Padvarninkai's origin.
Bukovanská and Ireland investigated the zircon grains in eucrites (they considered Padvarninkai brecciated). These zircon-bearing eucrites were classified according to the Zr/Hf ratio which showed a correlation with the degree of fractionation. Juvinas and Bereba show the least rations, Jonzac, Milbillillie and Cachari the middle ones, and Padvarninkai, Stannern and Pasamonte the highest. Zircon ages in Padvarninkai were 4553±13 Ma, synchronous that in Stannern [46]. However note that in composition Padvarninkai does not show too much similarity even with Pasamonte and/or Stannern. On maps 1-10 Pasamonte and Stannern are generally within the eucrite group while Padvarninkai is not; and Pasamonte is rather low in alkalies for an eucrite [12]. In 1993 Yamaguchi reported about the shock texture in Padvarninkai [47]. In 1994 Kunz reported about the argon ages in Padvarninkai [1]. In 1995 Shukolyukov and Begemann reported Pu-Xe datings for Padvarninkai, 5 other eucrites and one howardite [48], while Yamaguchi partially reconstructed the heat history of the hypothetical eucrite parent body [49]. 14 eucrites seem to show traces of a thermal metamorphism of prolonged duration and 720-880 °C temperature, definitely after differentiation and formation of basaltic crust. (May have it been the splintering impact of Vesta?)
Padvarninkai data accumulated enough to 1996 to draw some picture about its early history, and this was done in Nyqvist's lecture [50]; see also [1], of course. He considers Padvarninkai the most shocked eucrite, undergone 45-60 Gpa shock pressure. They traced events in 39Ar-40Ar, Rb-Sr and Sm-Nd data. The investigated sample was a grey gabbroic host with black, suevitic glass veins, impact-melted.
The argon age spectrum showed a maximum at 3.88±0.01 Ga, maybe corresponding to the eucritic clasts of age 3.80 Ga reported by Kunz & al [1]; and something is seen also about 1.7 Ga, reported also there. In the same time the Rb-Sr analyses gave 2 other pyroxene ages; one about 4.1 Ga, but with great error, and one quite recent, 1.15±0.12 Ga.
The Sm-Nd age was 4.51±0.11 Ga; from Nd isotope ratios and earlier Pu-Xe data it was possible to reconstruct somewhat the first cca 60 Ma of Padvarninkai after primary crystallisation at 4553±13 Ma (from the zircons).
According to Ref. [1], for any case, the histories of all the 3 lithologies of Padvarninkai were different. Lithology Pad-1, represented by fine-to-coarse-grained eucritic clasts, seems to have underwent a high-temperature stage with degassing at cca. 3.8 Ga (see above) and something happened with it (on more moderate temperature) also at 600 Ma. Pad-2, a fine-grained lithology with quenched texture and with the maskelynite of bytownitic composition had a serious event at 1.71 Ga (again, see above), but did not exchange Ar with the other lithologies. Pad-3, a partly glassy matrix, seems to have had event(s) between 3.71 and 3.44 Ga, and maybe melted at 600 Ma. So, they conclude that 600 Ma seems the "upper limit for the compaction age" of Padvarninkai.
The state of art in 1997 about the HED parent body reached to the age of parent eucrite basalt between 4.48 and 4.43 Ga [51] (see also [52]). The lecture compares the ratios e53 and 55Mn/52Cr for eucrites EET90020, Y792510 and Padvarninkai. The first ratio is 0.76±0.22, 0.789±0.19 and 0.64±0.13, respectively; the second is 2.99±0.15, 2.38±0.12 and 2.00±0.12. While the first ones are similar, Mn/Cr ones differ beyond 3 s. What is more, Mn/Cr=0.62 for the reconstructed HED parent body [51].
The lecture claims that Mn/Cr is enriched in the investigated 3 eucrites; certainly. One may tell that the much deviating 55Mn/52Cr values do not support the origin from the eucrite parent body; but note that Padvarninkai does not show too much similarity with even Y792510. The latter seems a quite familiar eucrite for bulk chemistry.
Investigations started to compare reflection spectra of special meteorites and asteroids; this is definitely hopeful to find parent bodies. Padvarninkai is among the selected meteorites, together with eucrites Stannern and Jonzac, a howardite and several chondrites [53]. We hope to hear about the comparisons soon; to be sure there are some specialities in the Padvarninkai spectra.
A recent paper explains Padvarninkai (together with a lot of other eucrites) via in-situ crystallization from a magma ocean on an asteroid [54]. Certainly such a process may have happened. The problem is the serious compositional difference between Padvarninkai on one hand [12], [21], [22] and all the other non-cumulate ones in the comparison. (Our maps contain Bereba, Cachari, Camel Donga, Ibitira, Jonzac, Juvinas, Pasamonte and Stannern of them.)
However, note that Chen and El Goresy raise doubts about the origin of maskelynite in general [16], [17]. In the traditional picture the precursor of maskelynite was never melted, but lost the crystalline structure in shock. In the suggested scenario the shock heated and melted the matter and then glass formed in the subsequent quenching. While both mechanisms are possible in a shock, the reconstruction of the impact, e.g. the value of the shock pressure will be different. And, for example, Shergotty, ALHA84001 &c. would not contain diaplectic glass but Chassigny would (cf. our Table 1). Such a revision would alter the reconstructed history of Padvarninkai as well, so let us wait for the new story. Maybe that will be clearer.
APPENDIX B: ON OUR COMPOSITIONAL DATABASE
We use a compilation of ours which is a homogenization of 4 compositional Tables + some individual datasets from literature. The large Tables are: the NIPR Catalog (YK) [10], the Jarosewich (J) compositional data [23], the Urey-Craig compilation (UC) [12] and the Wiik (W) measurements [24]. In addition we collected some interesting meteorites (S) from the literature, of which 10 are eucrites, 6 are shergottites and 3 are lunar basalts, namely
A-881394
EET-87548
LEW-86002
LEW-87010
PCA91007
RKPA8020
RKPA8022
Y-791186
Y-791195
Y-82037
ALH77005
ALH84001 (really an orthopyroxenite)
EET-79001
LEW-88516
QUE94201
Y-793605
A-881757
E-87521
Y-793169
Of the 968 metyeorites of our compilation 143 fall into the groups relevant now.
The different Tables cannot be simply united. We followed mainly YK, but YK did not give C, while J did it generally and the others occasionally. YK and J gave Fe2O3 as a rule; W did not, and UC generally not but sometimes did. For H2O YK and J strictly distinguished H2O+ and H2O-; W sometimes did it, sometimes not and UC generally not. There were also differences in handling S (FeS, S°, SO2 &c.), Ni (Ni°, NiO, NiS) and Co (Co°, CoO, &c.) data. We homogenized the forms. Now we can observe slight differences in trends between YK and J, and somewhat larger ones between them and the older Tables; we started to remove the systematic differences but here the uncorrected data are used.
Our dataset contains 968 measurements; a few meteorites are doubly represented and Serra de Mage trebly, but that is not an error from statistical point of view. This means 21 diogenites, 21 howardites, 79 eucrites (18 polymict), 14 shergottites and 7 lunar basalt meteorites, + Padvarninkai, shergottite in the UC Tables. The chemical constituents are as follows:
1) Oxides:
Si, Ti; Al, Cr, Fe(III); Mg, Ca, Fe(II), Ni, Mn; Na, K, P H(+), H(-)
2) Sulphide:
Fe(S)
3) Without bond:
Fe°, Ni°, Co, S, C
APPENDIX C: ON STATISTICAL ERRORS
The overall measurement errors can be estimated from the original Tables. YK and W are highly homogeneous in themselves; the majority of measurements were made by the same persons. It is true for the minority of measurements in J, but the same person was somehow included into the measurement or evaluation of the remainder too. UC is inhomogeneous, but the compilers checked the data in various ways.
Measurements generally give a total, which is not 100 %. The deviation has two main reasons. First, some other constituents are not listed. Second, each constituent is measured with some error.
Now, let i denote the constituent. The quantities xi are measured with errors dxi. Let us neglect correlation and let us use the Theorem of Central Limiting Distribution. Then we expect a Gaussian distribution for the total, x, with an average somewhat below 1 (or 100 %) and with a mean deviation
s2 = ĺi(dxi2) (C.1)
s obviously characterizes the overall quality of the measurements performed in thew same laboratory by the same group, &c.
If s is known, we may roughly estimate the mean errors for individual components on statistical grounds as
si » s/(xi)1/2 (C.2)
Now, it turns out that in all of YK, J, UC and W the large majority of points are within 3s from the averages of Gaussians, with only a few outlying measurements. Leaving out the outlying individuals, the self-consistent fits are
Table |
YK |
J |
UC |
W |
Average,% |
100.108 |
99.895 |
99.946 |
100.000 |
s,% |
0.342 |
0.396 |
0.619 |
0.962 |
Table 2: Error estimations
For the outlying measurements we assume that they are at their individual 1s‘s. For the compilation S we did the same and used ad hoc error estimations. Details will be given elsewhere.
REFERENCES
[1] J. Kunz, M. Trieloff, M. Bukovanská & E. K. Jessberger: 40Ar-39Ar Ages of Different Lithologies from the Unique Eucrite Padvarninkai. Lecture at the 57th Meeting of the Meteoritical Society, Prague, Bohemia, 28th July, 1994. Text: J. Kunz, M. Trieloff, M. Bukovanská & E. K. Jessberger: 40Ar-39Ar Ages of Different Lithologies from the Unique Eucrite Padvarninkai. Meteoritics 29, 487 (1994)
[2] Ch. Meyer: Mars Meteorite Compendium-1998. JSC #27672 (1998)
[3] J. N. Grossmann: The Meteoritical Bulletin N° 84, 2000 August (see in Meteoritics & Plan. Sci. 35, 199 (2000))
[4] A. Rubin & al.: The Petrology of Los Angeles..., Lecture at LPSC 31 (2000), Houston, March 13; text at http:/www.lpi.usra.edu/meetings/lpsc2000/pdf/1963.pdf
[5] P. Warren & al.: Geochemistry of Los Angeles..., Lecture at LPSC 31 (2000), Houston, March 13; text at http:/www.lpi.usra.edu/meetings/lpsc2000/pdf/2001.pdf
[6] J. Greenwood & al.: Late-Stage Crystallization Features of Los Angeles..., LPSC 31, (2000) Houston, March 13
[7] J. N. Grossmann & Jutta Zipfel: The Meteoritical Bulletin N° 85, 2001 July; to appear in Meteoritics and Planetary Science Vol. 36 (2001)
[8] R. Baalke: New Mars Meteorite Found in Oman. Dhofar 019. http://www.jpl.nasa.gow/snc/news39.html
[9] R. Baalke: Two New Mars Meteorites From the Sahara Desert. http://www.jpl.nasa.gow/snc/news37.html
[10] K. Yanai & H. Kojima: Catalog of the Antarctic Meteorites. NIPR, Tokyo, 1995
[11] R. L. Korotev: List of Lunar Meteorites (update: 01-Nov-2001). wysiwyg://48/http://epsc.wustl.edu/admin/resources/meteorites/moon_meteorites_list.html
[12] H. C. Urey & H. Craig: The Composition of Stone Meteorites and the Origin of the Meteorites. Acta Geochim. & Cosmochim. 4, 36 (1953)
[13] G. Tschermak: Die Meteoriten von Shergotty und Gopalpur. Sitzber. Math.-Naturw. Klasse Akad. Wiss. Wien 65, 122 (1872)
[14] H. Y. McSween & E. M. Stolper: Basaltic Meteorites. Sci. Amer. 242, 54 (1980)
[15] E. M. Stolper & H. Y. McSween: Petrology and Origin of the Shergottite Meteorites. Geochim. Cosmochim. Acta 43, 1475 (1979)
[16] M. Chen & A. El Goresy: The Nature of "Maskelynite" in Shocked Meteorites: Not Diaplectic Glass But a Glass Quenched from Shock-Induced Dense Melt at High-Pressures. Lecture at the 62nd Annual Meteoritical Society Meeting, Johannesburg, South Africa, July 11-15, 1999; www.lpi.usra.edu/meetings/metsoc99/pdf/5047.pdf
[17] M. Chen & A. El Goresy: The Nature of Maskelynite in Shocked Meteorites: Not Diaplectic Glass But a Glass Quenched from Shock-Induced Dense Melt at High Pressures. Earth and Planetary Science Letters 179, 489 (2000)
[18] D. Stöffler: Maskelynite Confirmed as Diaplectic Glass: Indication for Peak Shock Pressure Below 45 GPa in All Martian Meteorites. Lecture at LPSC 31 (2000), Houston, March 13; text at http:/www.lpi.usra.edu/meetings/lpsc2000/pdf/1170.pdf
[19] R. A. Binns: Stony Meteorites bearing Maskelynite. Nature 214, 1111 (1967)
[20] D. W. Mittlefehldt: in ALH84001 Petrographic Description. Antarctic Met. Newsletter 16, (3) 1993; http://curator.jsc.nasa.gov/curator/antmet/marsmets/alh84001/petdes.htm
[21] K. Slezevicius, M. Kaveckis & B. Kodaitis: Meteoritas, krites Lietuvoje 1929 metu vasario men. 9 diena. Kaunas Lietuvos Univ. Matematikos gamtos 5, 131 (1930)
[22] R. Brauns: Einige Bemerkungen zu dem in Litauen niedergefallenen Meteoriten. Centralblatt f. Miner. Geol. und Paläont. 5A, 401 (1930)
[23] E. Jarosewich: Chemical Analyses of Meteorites: A Compilation of Stony and Iron Meteorite Analyses. Meteoritics 25, 323 (1990)
[24] H. B. Wiik: The Chemical Compositions of Some Stony Meteorites. Geochim. Cosmochim. Acta 9, 279 (1956)
[25] Sz. Bérczi, Kamilla Gál-Sólymos, Ágnes Holba, B. Lukács & K. Martinás: On the Thermodynamics of Meteorites and Parent Bodies II: From Chondrites to Achondrites through the Primitive Achondrite Varieties (Stage A and Stage B) to the Basaltic Achondrites. Acta Mineral. Petrographica XL, 175 (1999)
[26] Sz. Bérczi & B. Lukács: Compositional Trends in Fe and Mg Contents of Chondrites. Lecture at Antarctic Meteorites, NIPR, Tokyo, June 10, 1997; text at Ant. Met. XXII, 6 (1997)
[27] Sz. Bérczi, Ágnes Holba & B. Lukács: On the Thermodynamics of Meteorites and Parent Bodies. KFKI-1996-15
[28] Sz. Bérczi, Ágnes Holba & B. Lukács: On the Thermodynamics of Meteorites and Parent Bodies: Portales Valley and the Borderland between Chondrites and Achondrites. KFKI-1999-01
[29] Sz. Bérczi & B. Lukács: Main Lines and Sidetracks in Basaltogenesis of Terrestrial Planetary Bodies. In: Side Tracks in Evolution, eds. Sz. Bérczi & B. Lukács, Proc. 7th Symp. on Matter Evolution, Dec. 6-7, 1996; KFKI-1998-07, p. 6
[30] Sz. Bérczi & B. Lukács: On Interplanetary and Interstellar Spherules. In: Proceedings of the International Meeting Spherules and Global Events, ed. by Cs. H. Detre & al., KFKI-1996-05, p. 113
[31] B. Lukács, Sz. Bérczi & Cs. H. Detre: Compositional Holes: Empty Places among Regions of Meteoritic and Planetary Materials on Various Compositional Fields. Lecture at TISS 1 Terrestrial Impacts & Spherules Symposium, Tokyo, June 13, 1997; Written text in TISS 1, ed. Y. Miura, p. 8
[32] K. Török & al.: Barium-Rich Hungarian Mesozoic Glassy Spherules with Quasiangritic Composition. Lecture at TISS 1 Terrestrial Impacts & Spherules Symposium, Tokyo, June 14, 1997; Written text in TISS 1, ed. Y. Miura, p. 30
[33] Nóra Fáy & B. Lukács: Cosmological Constraints for the Composition of Extrasolar Spherules. Spherula 1, 117 (1997)
[34] Sz. Bérczi, B. Lukács, A. Holba, A. Kiss & É. Papp: From FeO Reduction to Percolation and Outflow of Iron: Thermal Evolution of Chondrite Parent Bodies. Acta Miner.-Petrogr. XXXIX, 87 (1998)
[35] N. Fáy, B. Lukács, K. Martinás & Sz. Bérczi: On Cosmic Spherule Composition: Gravitation + Thermodynamics. Lecture at PIECE'99, Yamaguchi, Sept. 28, 1999. Text: PIECE'99, ed. Y. Miura, Yamaguchi, 1999, p. 25
[36] B. Lukács & Sz. Bérczi: Statistical Analysis of NIPR Meteorite Compositions, II: Comparison of Sequences of Differentiated Rocks from an Asteroid Sized Body and Earth. Lecture at Antarctic Meteorites, NIPR, Tokyo, June 11, 1997. Text at Ant. Met. XXII, 94 (1997)
[37] Sz. Bérczi, K. Brezsnyánszky, Cs. Detre, Z. Ditrói-Puskás, Nóra Fáy, Ágnes Holba, S. Józsa, I. Kubovics, B. Lukács, Gy. Szakmány & I. Tóth: High Titanium Basalts in the Solar System. Lecture at Antarctic Meteorites, NIPR, Tokyo, June 11, 1997. Text at Ant. Met. XXII, 9 (1997)
[38] G. Dreibus & H. Wänke: The Bulk Composition of the Eucrite Parent Asteroid and Its Bearing on Planetary Evolution. Z. Naturforsch. 35a, 204 (1980)
[39] K. Yanai: Unbrecciated Holocrystalline Meteorite 1152: Impactite or Crystalline Achondrite with Howardite Compositions? Lecture at PIECE'99, Yamaguchi, Sept. 28, 1999. Text: PIECE'99, ed. Y. Miura, Yamaguchi, 1999, p. 86
[40] P. Solt, Kamilla Gál-Sólymos, Sz. Bérczi & B. Lukács: New Investigations and Results on Kaba CV3 Carbonaceous Chondrite Texture, Reambulation of Its Spherules, and H2O-Na O Competition from NIPR Statistical Dataset. In: Terrestrial and Cosmic Spherules, ed. Cs. H. Detre, Akadémiai, Budapest, 2000, p. 155
[41] Eva Novotny: Introduction to Stellar Atmospheres and Interiors. Oxford University Press, New York, 1973
[42] Sz. Bérczi & B. Lukács: Statistics of Na Data and the NIPR Catalog. Poster at PIECE'99, Yamaguchi, Sept. 29, 1999. Text: PIECE'99, ed. Y. Miura, Yamaguchi, 1999, p. 15
[43] G. Tschermak: Die Bildung der Meteoriten und der Vulcanismus. Sitz. Kaiserl. Akad. Wien, Math.-Naturw. Klasse 71, 661 (1876)
[44] Sir Richard Ball: The Story of the Heavens. Cassel & Co., London, 1905
[45] M. J. Gaffey, J. F. Bell & D. P. Cruikshank: Reflectance Spectroscopy and Asteroid Surface Mineralogy. Asteroids II, (eds. R. P. Binzel & al.), p. 98. University of Arizona Press, Tucson, 1989
[46] M. Bukovanská & T. R. Ireland: Dating and Zr/Hf-Ratio of Zircons and Baddeleyites in Eucrites. Lecture at the 24th Annual Meeting of the AAS Division for Planetary Sciences Conf., Munich, Germany, 12th Oct., 1992
[47] A. Yamaguchi, H. Mori & H. Takeda: Mineralogy and Shock Texture in the Padvarninkai Eucrite. Lecture at the 56th Meeting of the Meteoritical Society, Vail, Colorado, 19th July, 1993. Text: A. Yamaguchi, H. Mori & H. Takeda: Mineralogy and Shock Texture in the Padvarninkai Eucrite. Meteoritics, 28, 462 (1993)
[48] A. Shokolyukov & F. Begemann: Pu-Xe Dating of Eucrites. Lecture at 26th Lunar and Planetary Science Conference, Houston, Texas, 13th March, 1995. Text: A. Shokolyukov & F. Begemann: Pu-Xe Dating of Eucrites. LPS XXVI, 1299 (1995)
[49] A. Yamaguchi, G. J. Taylor, H. Takeda & K. Keil: Global Thermal Metamorphism on the Eucrite Parent Body. Lecture at 26th Lunar and Planetary Science Conference, Houston, Texas, 13th March, 1995.
[50] L. E. Nyquist, D. Bogard, H. Wiesmann, C.-Y. Shih, A. Yamaguchi & H. Takeda: Early History of the Padvarninkai Eucrite. Lecture at the 59th Meeting of the Meteoritical Society, Berlin, Germany, 22nd July, 1996
[51] L. E. Nyquist, H. Wiesmann, Y. Reese, C.-Y. Shih & L. E. Borg: Sm-Nd Age and Mn-Cr Systematics of Eucrite EET90020. Lecture at the 60th Meeting of the Meteoritical Society, Maui, Hawaii, 23rd July, 1997; text at www.lpi.usra.edu/meetings/metsoc97/pdf/5220.pdf
[52] L. E. Nyquist, D. Bogart, H. Wiesmann and C.-Y. Shih: Meteor. &. Planet. Sci. 31, A101 (1996)
[53] M. Bukovanská, I. Nemec & M. Solc: Study of Some Achondrites and Chondrites by Fourier Transform Infrared Microspectroscopy and Drifts. Poster at the 61st Meeting of the Meteoritical Society, Dublin, Ireland, 27th July, 1998; text at www.lpi.usra.edu/meetings/metsoc98/pdf/full05.pdf
[54] J. A. Barrat, J. Blichert-Toft, Ph. Gillet & F. Keller: The Differentiation of Eucrites: The Role of In Situ Crystallization. Meteoritics & Planetary Science 35, 1087 (2000)
My HomePage, with some other studies, if you are curious.