Recent Scientific Papers on ALH 84001 Explained, with Insightful and Totally Objective Commentaries Allan Treiman Lunar and Planetary Institute Many scientific papers have now been published on the possibility that the martian meteorite ALH 84001 contains traces of ancient martian life (McKay et al. 1996a). Many (probably most) of these papers are difficult to understand (even for specialists), and many do not really say why they are important. Here, I've tried to present the main arguments of these papers for the educated nonspecialist, and some sense of why they are important (or why not). New this year, the first paragraph(s) after the title are a quick summary of the results (or executive summary, or sound bite). The more extended description follows in indented normal type. Last are my insightful and totally objective commentaries, worth exactly what you paid for them. My new year's resolution is to be more like Oscar the Grouch (of Sesame Street, if you've never had children). As before, the papers on ALH 84001 are given in reverse chronological order of publication date. --------------------------------------------------------------------------------- Jull A.J.T., Courtney C., Jeffrey D.A., and Beck J.W. (1998) Isotopic evidence for a terrestrial source of organic compounds found in Martian meteorites Allan Hills 84001 and Elephant Moraine 79001. Science 279, 366-369. Jull and co-workers measured the abundances of stable and radioactive isotopes of carbon in ALH 84001. Most of carbon in ALH 84001 is from its carbonate mineral globules (as reported previously). Most of the remaining carbon is from Earth organic material, i.e., terrestrial contamination. A small fraction of the carbon (~ 8%) is too old to be Earth contamination, and is not (in chemistry and carbon isotopes) like carbon from the carbonate minerals. This carbon may be from organic material formed on Mars, or possibly a rare inorganic mineral (also from Mars). Part of McKay et al.'s (1996) argument for traces of martian life in ALH 84001 is that the meteorite contains organic material, rich in PAH compounds, associated with its carbonate mineral globules. However, Becker et al. (1996) argued that this organic material is actually terrestrial contamination. To help resolve this issue, Jull and co-workers analyzed the isotopic composition of the carbon in the organic matter and the carbonate minerals of ALH 84001 (following Jull et al., 1997). The principal clue used by Jull is the abundance of the radioactive isotope of carbon, carbon-14, in the organic material. Carbon-14 is used as an age-dating tool for archaeological and cultural artifacts (like the Shroud of Turin). Carbon-14 forms continuously and abundantly in the Earth's atmosphere. As soon as a carbon-bearing compound is isolated from the atmosphere (e.g., a tree dies and stops absorbing CO2 from the air), its carbon-14 starts decaying away with a half-life of 5730 years. Most of the organic matter in ALH 84001 contains significant amounts of carbon-14 -- which means that it is terrestrial contamination (there is no reasonable extraterrestrial source of so much carbon-14). Also, the carbon-14 gives an average age near 6000 years, which is approximately 7000 years after ALH 84001 fell to Earth. So, there is little doubt that most of the organic carbon in ALH 84001 is terrestrial contamination. In addition, the relative abundances of carbon-12 and carbon-13 (the d 13C value) in the ALH 84001 organics are typical or carbon from living things on Earth. The carbon in carbonate minerals in ALH 84001 is clearly not terrestrial -- it has little or no carbon-14, and a d 13C value much higher than typical of Earth carbonates. Earlier, Jull et al. (1997) got similar results for carbonate minerals in a different sample of ALH 84001, although that sample had enough carbon-14 to suggest some chemical exchanges with Earth water. However, a small part of the carbon in ALH 84001 might be martian organic material. This carbon was not dissolved away during acid treatment designed to remove carbonate minerals, so it is either organic or some (unknown) resistant mineral. This batch of carbon has no carbon-14, meaning that it is very old. Jull and coworkers take this ancient age to mean that this batch of carbon did not form on Earth -- it is martian. This work appears to be carefully done, adequately documented, and carefully presented. It does not directly refute McKay et al.'s (1996) hypothesis of martian biological activity in ALH 84001, but it is not much of a confirmation, either. I have two comments about this work and possible evidence of martian biological activity in ALH 84001. ALH 84001 contains hundreds of parts per million organic carbon, much more than other martian meteorites except EETA79001 (which Jull also analyzed in this paper). This high abundance of organic matter has been used to support claims of fossil martian biology in ALH 84001. However, ALH 84001 contains the same amount of organic carbon as do typical basalt meteorites from asteroids, even those found in Antarctica (Grady et al., 1997). Just as Jull and co-workers showed that most of the organic carbon in ALH 84001 is terrestrial contamination, Grady et al. (1997) showed that most of the carbon in asteroidal basalt meteorites is terrestrial contamination. In this way, ALH 84001 is quite average and was not contaminated any more than normal for a meteorite. The most intriguing part of Jull's work, at least to me, is the extraterrestrial organic (?) material they found in ALH 84001. They found this carbon in a sample of ALH 84001 that had been treated to remove all of its carbonate minerals. At lower temperatures (<450^ÃC) this treated sample released the same terrestrial carbon (both 14C and d 13C) as the untreated sample. But at higher temperatures, the treated sample released some carbon without any 14C, meaning it was pre-terrestrial. This high-temperature, non-carbonate carbon could be organic matter, or could possibly be a rare, acid-resistant, as-yet-unidentified inorganic mineral. Many different kinds of organics can be released at these higher temperatures, including material like kerogen, graphite, and PAHs. So, it is tempting to say that Jull and co-workers detected the same PAHs that McKay et al. found (and probably also other high-temperature organic compounds). But most meteorite basalts from asteroids also contain about similar amounts of high-temperature carbon (10-30 parts per million; Grady et al. 1997). Could it be that basalts in the solar system just have this much of high-temperature carbon compounds, whether or not life was present? Citations: Becker L., Glavin D.P., and Bada J.L. (1997) Polycycic aromatic hydrocarbons (PAHs) in Antarctic Martian meteorites, carbonaceous chondrites, and polar ice. Geochim. Cosmochim. Acta 61, 475-481. Grady M.M., Wright I.P., and Pillinger C.T. (1997) Carbon in howardite, eucrite, and diogenite basaltic achondrites. Meteoritics Planet. Sci. 32, 863-87 Jull A.J.T., Eastoe C.J., and Cloudt S. (1997) Isotopic composition of carbonates in SNC meteorites, Allan Hills 84001 and Zagami. Jour. Geophys. Res. 102, 1663-1669. --------------------------------------------------------------------------------- Bada, J.L., Glavin D.P., McDonald G.D., and Becker L. (1998) A search for endogenous amino acids in martian meteorite ALH84001. Science 279, 362- 365. Bada and co-workers analyzed ALH 84001 for amino acids, chemicals that are essential in life as-we-know-it on Earth. In the meteorite's carbonate globules are small amounts of amino acids, which are nearly identical (in proportions of acid species and in their chemical handedness) to amino acids in Antarctic ice. So, Bada and co-workers conclude that (essentially) all of the amino acids in ALH 84001 are terrestrial contamination, carried into the meteorite by melted Antarctic ice. Part of McKay et al.'s (1996) argument for traces of martian life in ALH 84001 is that the meteorite contains organic material mixed with its carbonate mineral globules. Last year, Bada's research group claimed the organic material is terrestrial contamination (Becker et al., 1996). Continuing this work, Bada and co-workers analyzed ALH 84001 and its carbonate minerals for amino acids. Amino acids are small organic molecules, the building blocks of proteins and enzymes in all living things on Earth. Earth life only uses a few of the many possible amino acids in fairly characteristic relative abundances, and only uses the L form of those amino acids. With these distinctive characters, amino acids are a sensitive test for Earth organic contamination in meteorites. To analyze for amino acids, Bada and co-workers used a very sensitive technique developed in their laboratory. McKay et al. suggested that the signs of ancient martian life were associated with carbonate minerals in ALH 84001, so Bada and co-workers used a chemical extraction to separate amino acids in the carbonate globules from those elsewhere. First, they rinsed the samples of ALH 84001 in distilled water, and that extracted no amino acids at all. Then, they reacted the samples with weak hydrochloric acid, which should dissolve the carbonate minerals in the rock and release any amino acids associated with them. This acid solution was dried, and part of it analyzed for free amino acids (those not chemically bound to anything else). Another part of the solution was dried and treated to liberate amino acids that were bound to other molecules (for example, this treatment would break proteins into their constituent amino acids). And finally, they analyzed the remainder of the meteorite that was not dissolved in acid (including the pyroxene and chromite mineral grains) for bound amino acids. Bada and co-workers found that the amino acids in ALH 84001 were most abundant as bound acids associated with the carbonate minerals. There were almost no amino acids in the distilled water wash, the acid-insoluble residue, or as free amino acids in the acid solution. The part of ALH 84001 that dissolved in acid contained about 10 parts per million total amino acids (almost all chemically bound), while the rest of the rock contained only 75 - 100 parts per billion of amino acids. The amino acids in ALH 84001 are almost exactly in the same proportion as in the Antarctic ice -- the proportions of DL-serine to glycine to L-alanine are approximately 3:3:1. In addition, there is a little D-alanine in Antarctic ice and in ALH 84001 [ed. note: possibly from micrometeorites in the ice?]. This similarity of terrestrial and ALH 84001 amino acids leaves little doubt that they are primarily terrestrial contamination, derived from amino acids in the ice that was around ALH 84001. The amino acids that Bada and co-workers found in ALH 84001 are from the Antarctic ice. But this fact is not a death-blow to the hypothesis of that ALH 84001 contains traces of ancient martian life (McKay et al. 1996). Despite an exuberant press release from Scripps Oceanographic Institution, Bada's work is not a conclusive test of McKay's hypothesis. McKay et al. (1996) did not talk about amino acids, so the absence of preterrestrial amino acids does not refute their hypothesis at all. Of course, if Bada and co-workers had found abundant preterrestrial amino acids, it would have been strong support for McKay et al.'s hypothesis. Two aspects of Bada's experiments are puzzling (although probably not very important). First, their acid treatment was designed to dissolve carbonate minerals, but it dissolved 20% of their carbonate-free sample of lunar rock. What actually dissolved from the lunar rock? Possibly feldspar? Could feldspar (or whatever) also have dissolved from ALH 84001, and would this change the conclusions? Second, their acid treatment seems to have increased the masses of their samples. For instance, sample 2 of ALH 84001 started at 463 milligrams, and ended up as 472.5 milligrams (text and Table 1). What is this extra mass? Could it be lab contamination that might carry amino acids? Citations: Becker L., Glavin D.P., and Bada J.L. (1997) Polycycic aromatic hydrocarb ons (PAHs) in Antarctic Martian meteorites, carbonaceous chondrites, and polar i ce. Geochim. Cosmochim. Acta 61, 475-481. --------------------------------------------------------------------------------- Bradley J.P., Harvey R.P., and McSween H.Y.Jr. (1997) No 'nanofossils' in martian meteorite orthopyroxenite. Nature 390, 454. McKay D.S., Gibson E.K.Jr., Thomas-Keprta K., and Vali H. (1997) Reply to "No 'nanofossils' in martian meteorite orthopyroxenite." Nature 390, 455-456. Bradley et al. claim that the possible nanofossils found by McKay et al. (1996) in martian meteorite ALH 84001 are actually irregularities in the surfaces of mineral grains. These irregularities were accentuated by the metal coating that had to be put on the samples for electron microscope examination. So, Bradley and co-workers reject the hypothesis that ALH 84001 carries nanofossils of ancient martian life. In response, McKay et al. say that they also found the same surface irregularities, and that they are not possible martian nanofossils. The metal coating on the samples did not interfere with their identification of objects as nanofossils, because they did control experiments on metal coatings and know what the coating does. (G.J. Taylor has posted a nice summary of these letters.) Bradley and co-workers examined fracture surfaces of ALH 84001 using nearly the same methods that McKay et al. (1996) used. They found sausage-shaped surface features, approximately 100 to 400 nanometers (billionths of a meter) long, that looked (to them) similar to the possible nanofossils in the McKay et al. (1996) paper and in later magazine articles and press briefings. Bradley found these sausage-shaped features on the carbonate minerals (as McKay's `nanofossils' were) and also on the host silicate minerals. By observing the sample from many angles (in their electron microscope), Bradley found that the 'sausages' were not sitting on the host minerals, but were actually ridges poking out of the host minerals. Bradley also did a few experiments on how the metal coating on the samples changes the shapes of surface features. They found (as have others) that metal coatings tend to make surface features look segmented (the thicker, the more segmented) -- an appearance that McKay's group had suggested once to reflect cell boundaries. McKay et al. respond that they have also seen ridges on minerals' surfaces that Bradley et al. found -- same sizes, shapes, and textures. McKay and co-workers suggest that the ridges are grains of clay minerals formed during `incipient' alteration of the host minerals. But these surface ridges, say McKay et al., are not the possible nanofossils they described in 1996 and subsequently. Their possible nanofossils differ from the Bradley ridges by not being parallel with each other, by intersecting with each other at distinct angles, by being curved, and by being rather isolated from each other. McKay and colleagues also dispute that their identifications of possible nanofossils (here and earlier) were compromised by metal coatings on the samples. They reiterate that they did control experiments on the effects of metal coatings, and that the nanofossil morphologies do not result from coating. McKay also notes that some of Bradley's samples were coated with gold alone, rather than a gold-palladium alloy, and that gold coatings are known to make larger artifacts (artificial structures) than are gold-palladium. This exchange focuses on two important issues about the possible martian nanofossils in ALH 84001: 1) how can you recognize a nanofossil, and 2) how does laboratory preparation change the surfaces of the samples. Unfortunately, short "correspondence and reply" tidbits (Nature NuggetsÆ) cannot carry enough scientific "meat" to resolve these issues. 1) How can you recognize that a shape in ALH 84001 is a martian nanofossil? In 1996, McKay et al. cited "...regularly shaped ovoid and elongate forms ranging from 20 to 100 nanometers in longest dimension" as possible nanofossils (their Figure 6 and Kerr, 1996). At their big NASA press conference, McKay and colleagues also presented an image of aligned sausage-shaped objects in a grid-formation as being possible nanofossils. Bradley et al. found features that matched these characteristics, and showed that they were not biological. Here, McKay et al. seem to have changed their definition of martian nanofossils. Nanofossils are still elongate and ovoid. Now, however: they do not appear in parallel, but display "intersecting alignments;" they are relatively isolated from each other; they are significantly curved (their Fig. 2c); and they are much larger, up to 750 nanometers long. With these new criteria, many of McKay's own objects may not qualify as nanofossils: the ovoids of Figure 6a in McKay et al. (1996); the famous segmented worm shape (Kerr, 1996); and the aligned sausage-shaped objects. 2) How does the metal coating (for electron microscopy) affect the surfaces of minerals in ALH 84001? This question has been argued, mostly in private, since McKay et al. (1996) was published. In other words, are some of the `nanofossils' in ALH 84001 completely artificial, made during metal coating, and completely irrelevant to life on Mars? Believable answers to these questions will only come from carefully controlled experiments, where fragments of ALH 84001 are coated with various thicknesses of different metals and alloys. Bradley et al. report that they did a few experiments in this program; McKay et al. report that they did a series of experiments on a different sample (a lunar glass). Unfortunately, neither set of experiments has been reported in any detail, and I am still not sure of what metal coatings (Au or Au/Pd) do to surface morphology at these very small sizes. Citations: Kerr R.A. (1996) Ancient life on Mars? Science 273, 864-866. --------------------------------------------------------------------------------- Murty S.V.S. and Mohapatra R.K. (1997) Nitrogen and heavy noble gases in ALH 84001: Signatures of ancient Martian atmosphere. Geochim. Cosmochim. Acta 61, 5417-5428. About 4.0 billion years ago, traces of noble gases and nitrogen from the martian atmosphere were trapped in ALH 84001. The isotopic compositions and relative abundances of the heavy noble gases xenon (Xe) and krypton (Kr) are similar to the present-day martian atmosphere. So, Marsí unusual Xe and Kr compositions and abundances were set earlier than 4.0 billion years ago. Argon trapped in ALH 84001 has less 40Ar from radioactive 40K (potassium) that Mars' present-day atmosphere, suggesting that it has continued to gain 40Ar over time [ed. note: e.g., by volcanic outgassing]. Nitrogen trapped in ALH 84001 has much less of the heavy isotope 15N, consistent with loss of the light isotope 14N (and other light-weight gases) from Mars' atmosphere over the last 4 billion years. The elemental and isotopic composition of the martian atmosphere has been a real puzzle. It is greatly depleted in the light stable isotopes of all gas elements, from hydrogen to xenon. For instance, the abundance ratios of light to heavy xenon isotopes (e.g., 128Xe/136Xe) are approximately 0.7 times that in sun (Zahnle, 1993). It is a mystery how and when the light-weight isotopes were removed, but a separate process must have acted for each element (Pepin, 1994). Any process strong enough to remove a lot of, say, 128Xe compared to 136Xe, would certainly have removed all of the lighter gaseous elements completely (like krypton, argon, and nitrogen). Similarly, any process capable of separating 36Ar from 38Ar to the extent seen in the martian atmosphere would have removed essentially all of its nitrogen. One way to help understand the martian atmosphere would be to learn how its composition has changed through time. Its present-day atmosphere (analyzed by Earth telescopes and the Viking landers) is the same as the atmosphere of 180 million years ago, as trapped in some martian meteorites (most notably EETA79001). Recognizing that ALH 84001 has retained noble gases (like argon) for 4.0 billion years, Murty and Mohapatra investigated whether it might contain trapped martian atmosphere from that time. They used standard techniques - separating the meteorite into its minerals by their density, heating the samples up in steps of 200^ÃC (or more) to 1600^ÃC, and collecting the gases given off by each sample in each temperature step. The gases were separated, and the isotopic composition of each element was measured with a mass spectrometer. Murty and Mohapatra found that ALH 84001 contains significant quantities of nitrogen, argon, krypton and xenon gases. Most gases (xenon, krypton, nitrogen and 36Argon) all were released by the samples at nearly the same temperatures, suggesting that they are from the same trapped atmosphere component. ALH 84001 contains a nitrogen component comparable to Mars `mantle' (the Chassigny meteorite) and a trapped component with d 15N ^Ä +85per mil; the current Mars atmosphere has d 15N ª +620 per mil. From the isotopic composition of the argon (in mineral and temperature and temperature separates), the authors estimate that the trapped gas has 40Ar/36Ar £ 1400, while the current Mars atmosphere has a value of 2400. The trapped gas in ALH 84001 has 14N/36Ar about 60 times the value for the current Mars atmosphere. The Kr and Xe isotope compositions of most of the trapped gas are similar to the current martian atmosphere, or current atmosphere as modified by groundwater processes. Murty and Mohapatra infer that this trapped gas component is a sample of the martian atmosphere from 4.0 billion years ago, the age when argon gas was last lost from ALH 84001. The ancient and modern atmospheres have similar isotopes and relative abundances of xenon and krypton (the heaviest noble gases), which means that the hydrodynamic escape processes that set these abundances (Pepin, 1994) were complete by 4.0 billion years ago. The higher 40Ar/36Ar in the current atmosphere reflects production of 40Ar from potassium over the history of Mars. And the decrease in 14N/36Ar may reflect loss of nitrogen (through sputtering) into space over the last 4.0 billion years. This work is not directly related to the "life in ALH 84001" folderol. It is part of the long-term effort to learn about Mars' ancient environments through clues in the martian meteorites. The noble gases and nitrogen hold great promise in unraveling the evolution of Mars' atmosphere, particularly why it is so thin now (surface pressure of ~ 1/200 that of Earth) and where its water has gone. But this work, no matter how good, is not likely to be the final word from ALH 84001. The uncertainty here is not from Murty and Mohapatra's analyses, but in the inherent variability of samples of ALH 84001 and the many assumptions that must be made to unravel the noble gas story. First, it appears that different samples of ALH 84001 contain different quantities, proportions, and isotope compositions of the noble gases and nitrogen. This is perhaps not too surprising, as the mineral proportions and chemical composition of ALH 84001 are rather variable, for instance potassium abundances (108 vs. 200 parts per million: Mittlefehldt, 1994; Dreibus et al., 1994). For the noble gases, this variability can appear as differences in the proportion of 40Ar that comes from radioactive potassium (this paper; Turner et al., 1997), and as differences in xenon isotope ratios (Fig. 9 of this paper vs. Fig. 2 of Swindle et al., 1995 and Fig. 3 of Miura et al. 1995). Variability like these in elemental and isotopic abundances suggests that the gases in ALH 84001 came from many different sources and were not mixed well. It will be possible, eventually, to sort out the different sources (or components) of gas; now, it seems to be a muddle. Second, interpretation of noble gas and nitrogen abundances is not simple, and relies on some (fairly complex) correction schemes and underlying assumptions. Different research groups have not treated their data the same way; so when their results appear in conflict, it may be difficult for a non-specialist (like me) to understand why. For instance, all groups so far have agreed that some of the argon in ALH 84001 comes from atmosphere trapped in the mineral grains. Turner et al. (1997) present evidence that this trapped gas is like argon from the Earth's atmosphere: 40Ar/36Ar = 295. Murty and Mohapatra infer that the trapped argon is ancient martian, with 40Ar/36Ar £ 1410. Miura et al. (1995) and Goswami et al. (1997) use the current martian atmosphere value of 40Ar/36Ar ª 2400. Swindle et al. (1995) do not infer a specific 40Ar/36Ar for the trapped component. Is each group justified, given their data and the intrinsic variability of ALH 84001, or have some (or all) of them made unjustified simplifications in their data treatment Citations: Dreibus G., Burghele A., Jochum K.P., Spettel B., Wlotzka F., and W^Ônke H. (1994) Chemical and mineral composition of ALH 84001: A martian orthopyroxenite (abstract). Meteoritics 29, 461. Goswami J.N., Sinha N., Murty S.V.S., Mohapatra R.K., and Clement C.J. (1997) Nuclear tracks and light noble gases in Allan Hills 84001: Pre-atmospheric size, fall characteristics, cosmic ray exposure duration, and formation age. Meteoritics Planet. Sci. 32, 91-96. Mittlefehldt D.W. (1994) Errata. Meteoritics 29, 900. Miura Y.N., Nagao K., Sugiura N., Sagawa H., and Matsubara K. (1995) Orthopyroxenite ALH84001 and shergottite ALHA77005: Additional evidence for a martian origin from noble gases. Geochim. Cosmochim. Acta 59, 2105-2113. Pepin R.O. (1994) Evolution of the martian atmosphere. Icarus 111, 289-304. Swindle T.D., Grier J.A., and Burkland M.K. (1995) Noble gases in orthopyroxenite ALH84001: A different kind of martian meteorite with an atmospheric signature. Geochim. Cosmochim. Acta 59, 793-801. Turner G., Knott S.F., Ash R.D., and Gilmour J.D. (1997) Ar-Ar chronology of the Martian meteorite ALH84001: Evidence for the timing of the early bombardment of Mars. Geochim. Cosmochim. Acta 61, 3835-3850. Zahnle K.J. (1993) Xenonological constraints on the impact erosion of the early martian atmosphere. Jour. Geophys. Res. 98, 10899-10913. --------------------------------------------------------------------------------- Greenwood J.P., Riciputi L.R., and McSween H.Y.Jr. (1997) Sulfide isotopic compositions in shergottites and ALH 84001, and possible implications for life on Mars. Geochim. Cosmochim. Acta 61, 4449-4453. The authors measured the abundance ratio of sulfur isotopes (34S/32S) in minerals of martian meteorites to see if the sulfur in ALH 84001 had been processed by sulfate-reducing bacteria, as implied by McKay et al. (1996). They found no evidence for the action of sulfate-reducing bacteria in ALH 84001, and so reject the McKay et al. (1996) hypothesis that ALH 84001 contains traces of ancient martian life. The element sulfur occurs as two stable (not radioactive) isotopes with masses of 32 and 34, 32S and 34S. Most sources of sulfur have abundance ratios of 34S/32S that are very similar to the average in the solar system. However, sulfur that has been processed by bacteria (or other life forms) can have distinctly different abundances of these isotopes. The greatest changes in S isotopes come from sulfate-reducing bacteria, which take sulfate ions (SO42-) from water and convert them to sulfide ions (S2-) in water or as solid sulfide minerals. Sulfate-reducing bacteria, when they have lots of sulfate in water around them, can form sulfide minerals with ~5% less 34S than the sulfate in the water. This difference is easily detected, and has been used (on Earth) as a guide to the action of these bacteria. To estimate the sulfur isotope ratio for bulk Mars, Greenwood et al. measured sulfur isotope ratios the martian basalt meteorites (Shergotty, Zagami, EETA79001, LEW88516, and QUE94201). The sulfur isotope ratios for these meteorites are within 0.3% of the solar system average. In ALH 84001, they first measured sulfur isotopes in millimeter-sized grains of pyrite (FeS2), which are not associated with the possible traces of ancient martian life (Gibson et al., 1996; but see Shearer et al., 1997). The pyrite had variable and slightly `heavier' sulfur than the other martian meteorites, with 34S/32S from approximately 0.2 to 0.75% larger than the solar system average; this agrees with earlier work of Shearer et al. (1996). Finally, they analyzed the sulfur-rich outer zone of a single carbonate globule from ALH 84001 iron sulfide minerals in the carbonate globules were claimed by McKay et al. (1996) to have formed through the action of martian biological organisms. The outer parts of the carbonate globules contain carbonate and oxide minerals in addition to the sulfides, so Greenwood et al. did not get so precise a result here as for the pure sulfide minerals. Also, they had to apply a small correction for pairs of oxygen atoms masquerading as sulfur. But the 34S/32S for the sulfide-rich region of the carbonate globule is identical to the non-biological pyrite in ALH 84001: 0.6% larger than the solar system average. The non-biological and possibly biological sulfide minerals in ALH 84001 have nearly identical 34S/32S ratios. Greenwood et al. take this similarity to suggest that sulfur (in the possibly biological sulfides) in the carbonate globules was not processed by sulfate-reducing bacteria that the McKay et al. (1996) hypothesis is wrong. Rather, they suggest that all the sulfides in ALH 84001 formed from a high-temperature fluid (too hot for life-as-we-know-it), probably generated by an asteroid impact onto Mars. The variations in sulfur isotope ratios suggest mixing of `light' and `heavy' sulfur, the former perhaps from igneous rocks, the latter perhaps from Mars' surface. This paper is much weaker than it could have been because the authors did not document their experiments adequately. The analyses of sulfur isotopes in the pure sulfide minerals (pyrite and pyrrhotite) seem superb; they follow carefully described procedures, are based on good standards, and are repeatable. But the analysis of sulfur isotopes in the carbonate globule, the critical analysis for evaluating the hypothesis of ancient martian life (McKay et al., 1996), will be suspect until Greenwood et al. document it fully. The problem with Greenwood's analysis for sulfur isotopes in the carbonate globule is that they did not analyze only sulfide minerals. Their instrument, an ion microprobe, shoots cesium ions at the sample, and collects ions from the sample that are sputtered off by the cesium. Sulfur come off as S2- ions, both as the `light' 32S2- and the `heavy' 34S2-. Two problems are possible when the sulfur is present as sulfides among other minerals, like carbonates and oxides. If the sulfide minerals are mixed with oxide and carbonate minerals, the ion 16O16O 2- might be formed in abundance (from the carbonates and oxides) and might pass as 32S2-, as both ions have the same masses and charges. If there were lots of 16O16O 2- passing for 32S2-, the sulfur would appear `lighter' than it really is. It is also possible that having sulfur-bearing minerals among other minerals influences the way that the sulfur sputters off the sample and into the analyzer. For instance, sulfur in sulfides mixed with carbonates and oxides might sputter more like a sulfate than a sulfide, and require a different correction procedure. Greenwood et al. were aware of these potential problems, and reported that they: 1) corrected for the presence of 16O16O 2- (less than 0.2% in their value of 34S/32S); and 2) did experiments to show that their sulfur isotope correction procedures gave consistent results for 34S/32S with or without admixed carbonates and oxides. But they gave no details on the 16O16O 2- correction, and no results for the experiments on mixtures. Since we cannot see the details of their corrections, and the results of their experiments, we are really asked to take on faith that Greenwood did both properly. Some scientists, trusting the authors implicitly, will take their work on faith. Others, who do not accept the conclusions of Greenwood et al., will point to these problems as cause for discounting the paper entirely. And those who wish to "trust, but verify" will merely be disappointed. Citations: Gibson E.K.Jr., McKay D.S., Thomas-Keprta K.L., and Romanek C.S. (1996) E valuating the evidence for past life on Mars (letter). Science 274 , 2125. Shearer C.K., Layne G.D., Papike J.J., and Spilde M.N. (1996) Sulfur isotope systematics in alteration assemblages in martian meteorite ALH 84001. Geochim. Cosmochim. Acta 60, 2921-2926. Shearer C.K., Spilde M.N., Wiedenbeck M., and Papike J.J. (1997) The petrogenetic relationship between carbonates and pyrite in martian meteorite ALH 84001 (abstract). Lunar Planet. Sci. XXVIII, 1293 -1294. --------------------------------------------------------------------------------- Turner G., Knott S.F., Ash R.D., and Gilmour J.D. (1997) Ar-Ar chronology of the martian meteorite ALH 84001: Evidence for the timing of the early bombardment of Mars. Geochim. Cosmochim. Acta 61, 3835-3850. The authors studied the age of ALH 84001, using 39Ar-40 Ar (argon-argon) radio-isotope dating. For the `traces of ancient life' controversy, their most important result is a revision of the 3.6 billion-year-old age that McKay et al. (1996) used as the time of carbonate formation. Turner et al. have revised the age for this particular sample of carbonate to 3.83 + 0.15 billion years, within uncertainty of nearly all other 39Ar-40Ar ages for ALH 84001. This `carbonate' age may not be when the carbonates formed. It actually is the age of the feldspar-composition glass that is mixed with the carbonate minerals, which could be older, younger, or the same as the carbonates. This paper represents an exhaustive study of the 39Ar-40Ar age of ALH 84001; this radioactive age-dating system is actually the potassium-argon (K-Ar) system, but some of the potassium is converted to 39Ar (in a nuclear reactor) so it can be measured at the same time as the 40Ar. Also, Turner and coworkers calculated the cosmic ray exposure age of ALH 84001 and its abundance of trapped martian atmosphere. Turner et al. studied three rock fragments by `stepped heating:' heating each sample up 100^ÃC at a time and collecting all the argon that was released at each temperature. This method allowed them to tell what abundances of argon isotopes were released by each kind of mineral in the fragment. Turner et al. also analyzed 40 spots on thin sections (microscope slides) by vaporizing them with a laser beam and collecting the argon that was released. After corrections for various sources of argon, including contamination from martian atmosphere, Turner et al. found that nearly all of the samples were consistent with an 39Ar-40Ar age of 3.97 billion years, possibly as old as 4.05 billion or as young as 3.8 billion. This 39Ar-40Ar age for ALH 84001 is essentially the same as determined by other research labs (Bogard and Garrison, 1997; Goswami et al., 1997). Two samples gave older ages, near 4.4 billion years; its is not clear if these ages are real. The age of sample 110i, rich in carbonate minerals, was originally reported as 3.6 billion years (Knott et al., 1996); McKay et al. (1996) took that as the age of the carbonate globules and their possible signs of Martian life in ALH 84001. This ancient age was important, as it placed the possible signs of martian life in the distant past, when Mars was probably much wetter (and possibly much warmer) than it is now. This ancient `warm, wet' Mars would have been similar to the ancient Earth, and so a reasonable place for life to form and flourish. However, Turner et al. have revised the age of sample 110i to 3.83 + 0.15 billion years, which is (within uncertainty) the same as nearly all other 39Ar-40Ar ages for ALH 84001. Further, sample 110i contains a LOT of potassium, much more than could have come from the carbonate minerals alone. The potassium in 110i probably came from silicate glass (like maskelynite) mixed with the carbonate, and so its 39Ar-40Ar age is the formation (or last heating event) of the silicate glass! So the age of spot 110i really does not limit the age when the carbonate formed. In calculating the 39Ar-40Ar ages of their samples, Turner et al. had to determine the proportion of Earth and martian atmospheres in their samples, and also how long the samples were exposed to cosmic rays in interplanetary space. Some samples had significant proportions of Earth atmosphere, but most had relatively little martian atmosphere. On average, less than 5% of the 40Ar in the samples came from the present-day martian atmosphere; this 40Ar probably was forced into the glass in ALH 84001 when it was ejected from Mars. That probably happened approximately 14 million years ago, the cosmic ray exposure age (see the paper below by Eugster et al., 1997). The 39Ar-40Ar age of approximately 4.0 billion years fits well with the ages of planetary bodies in the solar system. Most rocks from the Moon's highlands give 39Ar-40Ar ages from 3.8 to 4.0 billion; the oldest rocks on Earth formed at about 4.0 billion; many meteorites were shocked by impact between 4.1 and 3.5 billion years ago. Turner et al. suggest that the 39Ar-40Ar age of ALH 84001 represents an asteroid impact onto Mars (Treiman, 1995), and that impact was approximately at the same time as the large impact basins formed on the Moon. This correspondence seems to support the idea of a `lunar cataclysm' at about 4.0 billion years ago - a time when the Moon's surface was especially hard hit by asteroids. This paper was submitted for publication in August, 1996, back when ALH 84001 was most interesting as a sample of the ancient Martian crust. That was the impetus for this study -- G. Turner and his group wanted to understand the age and impact history Mars, especially as it might relate to the Moon. Ar-Ar ages for moon rocks cluster at 4.1 - 3.8 billion years ago, which suggests to some people that this was a time of abundant large asteroid impacts on the Moon - the so-called `lunar cataclysm.' Other people have concluded that the Moon was hammered by asteroid collisions continuously from 4.5 billion years ago through 3.8 billion, but that older ages were erased by younger ones. The results here seem consistent with the notion of an impact `cataclysm' happening throughout the inner solar system. For this study, Turner and colleagues had to know which event was actually being dated by the Ar-Ar system, they accepted Treiman's (1995) history as best fitting their data. Treiman (1995) proposed that ALH 84001 experienced two shock events: one that granulated and sheared the rock, and a second (after the carbonate globules formed) that produced shock glass with little deformation. Turner et al. assigned their age to the earlier event, and noted that production of shock glasses commonly does not reset Ar-Ar ages. However, Turner et al. did not consider more recent, alternate histories for ALH 84001; they were proposed after this paper was written. Bradley et al. (1996, 1997) gave evidence that ALH 84001 was heated to above 500^ÃC during formation of some magnetite grains, and (they infer) during formation of the carbonate globules. Scott et al. (1997) inferred that the carbonate globules and the feldspar-composition glass formed simultaneously in a single shock event. If either of these scenarios were true, they would most likely be recorded by the Ar-Ar age dates and could have happened 4.0 billion years ago. Gleason J.D., Kring D.A., Hill D.H., and Boynton W.V. (1997) Petrography and bulk chemistry of Martian orthopyroxenite ALH 84001: Implications for the origin of secondary carbonates. Geochim. Cosmochim. Acta 61, 3503-3512. Gleason and coworkers did a general study of ALH 84001, emphasizing microscope observations and chemical compositions of the rock and its minerals. Particularly, they examined the carbonate globules which McKay et al. (1996) suggested were formed by ancient martian life. Gleason and coworkers infer that the globules were deposited from liquid water, and so disagree with Harvey and McSween (1996) and Scott et al. (1997), who claimed that the carbonate minerals formed at high temperatures from molten carbonates. However, Gleason saw no evidence that the carbonate globules were associated with life, and so do not support McKay et al. (1996). On the contrary, they noted that similar carbonate globules have formed in other meteorites and on Earth without any apparent biological influences. Gleason and coworkers inferred, from mineral textures, that the carbonate globules grew from water-rich fluid cooler that 300^ÃC. The carbonate globules appear to have formed by replacing material with the composition of plagioclase feldspar. Treiman (1995) had inferred that this material was crystalline feldspar, but Gleason noted that carbonate replacing crystalline feldspar grows as crack filling and veinlets, NOT as globules. So, they conclude that the carbonates in ALH 84001 replaced feldspar glass, not crystals. If the feldspar glass ever been hotter than 300^ÃC for a few hours even, it would have crystallized back to plagioclase again. Gleason inferred that the this feldspar glass formed at the same time as did the granular bands (crush zones) that criss-cross the meteorite. Gleason and co-workers also observed is that the chemical composition of ALH 84001 varies a bit. They analyzed the chemical composition of two 1/3-gram fragments from different parts of ALH 84001. Some elements (like lanthanum) are five times less abundant in the fragment from a `crush zone' than the other fragment. Similar variability is apparent in other published chemical analyses. Gleason thinks this variability arose as some elements (like lanthanum) moved around in ALH 84001 before the carbonate globules grew. Finally, Gleason noted that pyrite, an iron sulfide mineral, was associated with chromite. They did not mention finding any pyrrhotite, another iron sulfide mineral. The significance of these observations is discussed below. Gleason and co-workers have provided a wealth of new chemical data on ALH 84001, and their excellent microscope observations (although important) do not resolve the issue of ancient life in ALH 84001. Rather, their work serves to emphasize the depth of disagreement about ALH 84001, and how much remains to be learned about the rock. As for the carbonate globules, Gleason and co-workers support the low-temperature position of Romanek et al. (1994), Treiman (1995) and Valley et al. (1997); low-temperature here means < 300^ÃC, which could still be much too hot for life as we know it. Gleason sees no evidence for the very high temperatures (> 500^ÃC) inferred by Harvey and McSween (1996), Bradley et al. (1996), and Scott et al. (1997). Gleason and co-workers do not have proof that the carbonates formed without life, just their reasoned judgment that life is not absolutely required to produce the structures and compositions they found. Their inference that the carbonate globules replaced glass rather than crystalline plagioclase is intriguing, and seems to be more realistic than my 1995 suggestion that the carbonates replaced crystalline plagioclase. However, there is no general agreement on how the carbonate globules formed; others have claimed that they replace pyroxenes or that they filled cracks and bubbles in the rock. The variability of the chemical composition of ALH 84001 is not surprising. ALH formed when crystals of the mineral orthopyroxene grew in a mass of basalt magma, and settled out to the bottom of the mass. Elements like lanthanum would have been concentrated in the magma among the settled crystals. So the amount of lanthanum in a piece of ALH 84001 would represent how much magma was caught among the orthopyroxene crystals. And the amount of magma might vary simply because the crystals were packed together tighter is some spots. On the other hand, the low-lanthanum sample is from a `crushed zone', and it is possible that the crushing managed to squeeze some lanthanum-bearing mineral (like plagioclase glass) out of that area. Finally, the observations here remind us of problems with the sulfide minerals in ALH 84001. First, Gleason and coworkers found pyrite (FeS2) associated with chromite rather than with the carbonate globules as reported by most other workers. The chromite has nothing to do with the hypothesis of fossil life in ALH 84001, while (of course) the carbonate globules do. Now, the sulfur isotope ratio (34S/32S) in the pyrite does not look those in Earth life, and so seemed to mean that the carbonate globules could not be associated with life (Shearer et al., 1996; Greenwood et al., 1997; Shearer, 1997; Shearer and Papike, 1996, 1997). However, if the pyrite did not form with the carbonate globules, its sulfur isotope ratio is not relevant to the hypothesis of life. Second, Gleason and co-workers did not mention finding any pyrrhotite (Fe1-xS) in ALH 84001; in fact, no pyrrhotite has been seen in thin sections. This absence is peculiar, as Kirschvink et al (1997) found that the magnetism in ALH 84001 was trapped in pyrrhotite! Where is the pyrrhotite, or could the magnetic signature be from some other mineral? Eugster O., Weigel A., and Polnau E. (1997) Ejection times of Martian meteorites. Geochim. Cosmochim. Acta 61, 2749-2757. The authors used abundances of `cosmogenic nuclides,' produced when a meteorite is exposed to cosmic rays, to measure how long four martian meteorites were in interplanetary space. ALH 84001 was exposed to cosmic rays for 14.4 + 0.7 million years, which probably is the time when ALH 84001 was blasted of f Mars. This cosmic ray exposure age for ALH 84001 is similar to ages found by other researchers (e.g., Goswami et al., 1997). None of the other martian meteorites were exposed in interplanetary space for so long, so it seems fairly certain that ALH 84001 did not come from the same site on Mars (impact crater on Mars) as any other martian meteorite. Hutchins K.S. and Jakosky B.M. (1997) Carbonates in martian meteorite ALH84001: A planetary perspective on formation temperature. Geophys. Res. Lett. 24, 819-822. The possible traces of life in ALH 84001 are all associated with its carbonate mineral globules, and so the formation of the globules is very important. If the globules formed was hotter than about 150^ÃC, a biological origin seems quite unlikely. Low formation temperatures, less than 80^ÃC, have been derived from the abundances of oxygen isotopes (16O and 18O) in the carbonates by Romanek et al. (1994). Here, Hutchins and Jakosky suggest that Romanek et al.ís temperature estimate was too low. Romanek et al. used oxygen isotope ratios as a thermometer, by comparing the oxygen isotope ratios (18O/16O) of a mineral and the liquid it grew from. The greater the difference in 18O/16O between the mineral and liquid, the lower the temperature would have been. None of the liquid is trapped in ALH 84001, so Romanek et al. had to estimate its oxygen isotope ratio as something like normal waters on Earth. Hutchins and Jakosky point out that oxygen and carbon in the martian atmosphere are much richer heavy isotopes of oxygen and carbon (18O and 13C) than in the Earthís atmosphere, and so Mars' water is also likely to have a relatively high 18O/16O and 13C/12C ratios. When put into the oxygen isotope thermometer, this difference means that the ALH 84001 carbonates probably formed between 40^ÃC and 250^ÃC. This paper emphasizes yet another uncertainty in determining the temperature of formation of the carbonate globules in ALH 84001. Whether Romanek et al. or Hutchins and Jakosky are more correct depends on two questions. First, did Mars' atmosphere have its current high 18O/16O and 13C/12C ratio before the carbonates formed? If not, Hutchins and Jakosky's argument is not valid. Today, Mars' atmosphere has a significantly higher 18O/16O and 13C/12C than martian rocks (the meteorites), and this difference means that the atmosphere somehow lost much of its original 16O and 12C to space. How the atmosphere lost these light isotopes is not certain, but Mars' low gravity (compared to Earth) and weaker magnetic field were probably important. When the light isotopes left Mars' atmosphere is not known (Jakosky and Jones, 1997); unfortunately, when the carbonates were deposited is not really known either. Second, did the liquid that deposited the carbonates come (eventually) from Mars' atmosphere? Hutchins and Jakosky's paper works from the idea that the liquid came from the atmosphere, and shared its high 18O/16O and 13C/12C ratios. But it is possible that the liquid came from deep inside Mars (in the jargon, 'juvenile water'), and never contacted the atmosphere. In that case, the high 18O/16O and 13C/12C of the ALH 84001 carbonates came entirely from a low formation temperature. --------------------------------------------------------------------------------- Scott E.R.D., Yamaguchi A. and Krot A.N. (1997) Petrological evidence for shock melting of carbonates in the martian meteorite ALH84001. Nature 387, 366-379. Here's a new theory of the origin of the carbonate globules in ALH 84001: they formed at very high temperature, during an asteroid impact on Mars, from carbonate rich melt. If the globules formed this way, they could not have been hosts to ancient martian life forms (McKay et al., 1996). The authors' argument is in four parts: 1) the clear silicate glass in ALH 84001 was melted during an impact shock (presumably an asteroid hitting Mars); 2) all the shock features in ALH 84001 formed in this same shock event; 3) the small dispersed grains of carbonate minerals were once molten, like the glass, because they all share similar structures and textures; and 4) and the structures and textures of the large carbonate globules also fit with once being molten. First, the authors show that the clear glass was once molten, a liquid. This glass had been called 'maskelynite,' which forms from feldspar minerals during shock without melting. The authors here show that the glass was molten because: its shapes were modified by shock, veinlets of the glass were injected into other minerals, it contains flow features, and it contains bubbles. Further, the chemical composition of the glass is not just the same as feldspar minerals; in addition to feldspar, the glass contains extra silica and sometimes extra chromium (from the mineral chromite). [The authors do not give a melting temperature, but it was much more than 1000^ÃC!] Second, the authors suggest that all the shock features in ALH 84001 formed in the same shock event that melted the glass. They note that single impact events can produce lots of different shock effects in a single rock, and the effects can cut across each other. They infer that the 'crush zones' or granular bands that criss-cross the rock were the first shock effect, and that the glass formed next [probably within seconds or a minute]. Third, the authors see that the glass and the small carbonate grains have similar shapes, and infer that both formed in the same way. The glass and small carbonates both enclose pyroxene grains in rounded shapes, and fill cracks in grains. Some cracks contain both carbonate minerals and the glass, which suggests to the authors that the cracks (and the "crush zones") formed at the same time as both the glass and the carbonates. So, the authors suggest that the glass and the carbonate melted at the same time and squirted into and around other minerals in ALH 84001. Carbonate melts are very runny, so they would squirt more easily into cracks; there is more carbonate than glass in the cracks. The authors looked for evidence in support of other proposed origins for the carbonates, and found none to their satisfaction. Fourth, the authors demonstrate that the carbonate globules could have formed as melt droplets, just like the small carbonate grains. The small carbonate grains cover the same range of chemical compositions as the large grains, suggesting that they formed at the same time in the same process (as impact melts). The shapes of the globules in the glass are like liquids that don't mix (like oil droplets in water); carbonate melts do not mix with silica-rich melts, and can form rounded shapes like the globules in ALH 84001. The authors also cite cases on Earth where carbonate minerals have been melted and moved around during impact shocks. So, all the carbonates now in ALH 84001 formed at very high temperatures. This theory is completely inconsistent with the inferences of McKay et al. (1996) that the carbonate globules contain evidence for martian life. ALH 84001 must have contained carbonate minerals before they were shock melted, but the origin of these ancestral carbonates is not known. In my opinion, this paper does not refute McKay et al. (1996), because it doesn't prove that the carbonate globules formed at a high temperature. The actual observations here are new and convincing, and it seems certain that that the clear glasses in ALH 84001 were once molten. There remain (to me) some stumbling blocks between this conclusion and the claim that all the carbonate globules formed at the same high temperature as the clear glass. The biggest doubt is whether the carbonate globules were ever molten, whether they actually were rapidly cooled droplets of carbonate melt. These observations, among others, seem difficult to explain if the carbonate globules were once molten. * The mineral grains in each carbonate globule grew outward toward the globule's rim. However, crystals growing from a melt globule will usually grow inward from the rim because melts crystallize as they cool down, and the rim of a globule is its first part to cool (like Fig. 15.6b of Kjarsgaard and Hamilton, 1989). * The carbonate globules have different chemical compositions inside and out (from brownish cores rich in calcium and iron to water-clear rims rich in magnesium). This zoning is unlikely from carbonate melt globules in two ways. If the globules cooled really fast, they ought to have little zoning because the calcium, iron and magnesium in the melt would grow into the crystals before the magnesium had time to separate from the calcium and iron. On the other hand, if cooling were slow enough that the calcium, iron, and magnesium in the melt could move around, the solidified globules ought to be zoned the other way: cores rich in magnesium and rims rich in calcium and iron (Scott et al. mention this problem).> There are other problems here too, and they will be explored at length. First, the evidence that all the shock features in ALH 84001 formed in a single impact event is not (to me) very convincing, compared to evidence for multiple impacts (Treiman, 1995). Second, McKay and Lofgren (1996) showed a picture of a the glass cutting across the Ca-Fe-Mg layering and 'oreo cookie' rim of a carbonate globule. This structure seems difficult to make if the glass and carbonate were liquid at the same time. And third, the zoning in oxygen isotopes from core to rim in the globules (Valley et al., 1997; Leshin et al., 1997; Saxton et al., 1997) may be impossible to produce at the high temperatures needed to melt these carbonates. For more about this paper, check out the University of Hawaii's Planetary Science Research Discoveries webzine. --------------------------------------------------------------------------------- Kirschvink, J. L., Maine A. T., and Vali H. (1997) Paleomagnetic evidence of a low-temperature origin of carbonate in the martian meteorite ALH 84001. Science, 275, 1629-1633. When a rock forms or cools down, it can trap some of the local magnetic field; magnetic minerals in the rock become little bar magnets, aligned with the planet's magnetic field. This trapped magnetic field, called natural remnant magnetism or NRM, can stay in the rock indefinitely, and can be used to unravel the history of the magnetic minerals and the rock. The strength of the trapped magnetic field can tell how strong the planet's field was. If the rock is broken or bent, the magnetic field trapped in it will point in a different direction from the original field. If the rock gets heated above a critical temperature, the old trapped magnetic field is lost, and a new one is trapped when it cools down again. For the McKay et al. (1996) hypothesis of fossil martian life in ALH 84001, the most important result from Kirschvink et al. is that the carbonate globules formed below 325^ÃC, and probably below ~110^ÃC. McKay et al. require a low formation temperature to permit bacterial growth, and many types of Earth bacteria and archaea can live and prosper at 110^ÃC! The upper temperature limit is too high for known Earth life, but is an UPPER LIMIT, and is still better for McKay et al. than the 500^Ã-700^ÃC temperatures estimated by other groups. The argument for carbonate formation below 325^ÃC is indirect, but fairly clear. Kirschvink et al. measured the trapped magnetic fields (NRM) in two adjacent fragments of ALH 84001 from the fracture zone where McKay et al. found the most carbonate globules. The trapped fields in the two fragments were strong, equally strong, but in different orientations; the "bar-magnets" of the magnetic minerals were pointed in different directions. This meant that the two fragments had probably trapped the same original field, but had been rotated or jostled when the fracture between them formed. If ALH 84001 had ever been hotter than 325^ÃC since the fragments were jostled, they would have lost their original magnetic fields; when they cooled, the fragments would have trapped the new magnetic field, with the same direction in both fragments! Because the fragments do have magnetic fields in different directions, ALH 84001 could not have been hotter that 325^ÃC at any time after the fractures formed. Now, the carbonate globules are in these same fractures, and must have formed after the fractures did, and so must not have formed at temperatures hotter than 325^ÃC (otherwise the rock fragments would have their trapped magnetic fields pointing in the same direction)! The argument for carbonate formation below ~110^ÃC depends on the details of how the trapped magnetic field changes as the rock is heated. In ALH 84001, the trapped magnetic field is in the iron sulfide mineral pyrrhotite. When pyrrhotite is heated to temperatures below its critical temperature of 325^ÃC, its trapped magnetic field fades away somewhat. But Kirschvink et al. found no hint of this fading in ALH 84001's trapped magnetic field. The 110^ÃC temperature actually comes from their sample preparation, not anything inside the rock. They had to heat their samples to 110^ÃC to allow their glue to cure. If ALH 84001 had been heated to >110^ÃC on Mars, any magnetic effects would have been erased as the glue cured. The results of this paper are a strong challenge to "anti-life in ALH 84001" scientists. However, the results are not (yet) proof of a low-temperature origin and certainly not proof of life on Mars. Although I am not an expert on magnetism, I see two issues in this work as it relates to McKay et al.'s hypothesis that ALH 84001 contains traces of ancient martian life. The first issue is the timing of fracturing of ALH 84001 compared to the timing of carbonate formation. ALH 84001 was fractured at least twice, before and after the carbonate globules formed. Many carbonate globules sit in fractures, so these fractures must have been there first (McKay et al., 1996). The carbonate globules are themselves sliced and broken along fractures, which must have come later (Mittlefehldt, 1994; Treiman, 1995; McKay et al., 1997). So, could Kirschvink's two rock fragments have separated by a late fracture, rather than an early fracture? If this particular fracture formed after the carbonate globules were deposited, Kirschvink's results here would say nothing about formation of the carbonate globules. The second issue is the absolute age of the carbonate globules, which should be ancient (3.6 billion years old) according to McKay et al. (1996). The problem here is that the tiny magnetite grains in the carbonate globules have not trapped any detectable magnetic field themselves. The magnetites do contribute to other magnetic properties of the rock, just not the trapped field (the NRM). Could this mean that the magnetite grains grew when there was no field, and so are fairly young (Wadhwa and Lugmair, 1996)? Or could it mean that Kirschvink's sample had so few carbonate globules that their trapped magnetic field could not be detected? The most important result from this paper, particularly for life on Mars, is the evidence that Mars had a strong magnetic field! Mars now has no detectable magnetic field, and had hardly any field 1.3 billion years ago, when many of the martian meteorites formed. Kirschvink et al. have demonstrated that Mars had a strong magnetic field (possibly as strong as the Earth's is now) about 4.0 billion years ago, when ALH 84001 cooled. First, a strong magnetic field would have protected Mars' surface from much deadly radiation from space. Its magnetic field would have deflected radiation like electrons and protons from the Sun, just as the Earth's magnetic field protects us now. Second, and perhaps more important, a magnetic field early in Mars' history would have protected its atmosphere. Mars' atmosphere is now quite thin, about 1/200 as thick as the Earth's. Without a thick atmosphere, Mars' surface could never have been warm enough to permit liquid water, and there is very good geologic evidence that liquid water was once abundant on the surface of Mars. What happened to Mars' atmosphere? Much of it was swept away by the solar wind, the continual stream of electron and protons that shoot off the Sun. But a strong magnetic field would have protected Mars' atmosphere, possibly letting Mars' surface be warm and wet enough for life to develop. --------------------------------------------------------------------------------- Valley J.W., Eiler J.M., Graham C.M., Gibson E.K.Jr., Romanek C.S., and Stolper E.M. (1997) Low-temperature carbonate concretions in the martian meteorite ALH 84001: Evidence from stable isotopes and mineralogy. Science, 275, 1633-1638. The temperature of formation of carbonate globules in ALH 84001 is important because the globules are hosts to the possible traces of ancient martian life (McKay et al., 1996). The first estimates of the globules' formation temperature, <320^ÃC, relied on oxygen isotope measurements (Romanek et al., 1994); here, Valley et al. revisit the oxygen isotope measurements with a new improved analytical method and confirm the low formation temperature. Valley et al. used an ion microprobe to determine oxygen isotope abundances in the carbonate globules and other minerals in ALH 84001. The ion microprobe can produce analyses from very small spots, about 20 micrometers (µm) in diameter, which is important because the carbonate globules are <200 µm in diameter. Valley et al. analyzed oxygen isotope ratios in carbonates from two separate ellipsoids, one of which was a composite of two smaller carbonate bodies. To help calibrate the ion microprobe measurements, Valley et al. also obtained chemical analyses of these and nearby spots in ALH 84001 using an electron microprobe. Valley's results are consistent with, and expand on, the earlier work of Romanek et al. (1994). They found that the carbonate minerals were variably enriched in the heavy oxygen isotope 18O, with enrichments ranging from d18O = 9.5 to 20.5 "per mil" (or parts per thousand). Carbonate near the globule rims was much richer in 18O than carbonate from the cores, and the different globules had different 18O enrichments in their cores. Valley et al. inferred that the carbonate globules formed at low temperatures because their chemical and isotopic variations could not have been preserved, if they had formed at high temperatures. Valley et al. estimate that the carbonate globules formed at <100^ÃC. An absolute upper temperature limit from their results comes from assuming that the carbonates were in oxygen isotopic equilibrium with the surrounding pyroxene. This upper limit on temperature is ~300^ÃC; the temperature had to have been lower because the pyroxene and carbonate were not in chemical equilibrium. Valley et al. also made some interesting discoveries and observations during their work: (1) They also analyzed the carbonates for carbon isotope composition, and found some evidence for an organic carbon component that has relatively little of the heavy carbon isotope 13C. This finding is one of a number of hints now of very "light," possibly organic, carbon in ALH 84001. (2) Valley et al. found a veinlet of silica that cut across one of the carbonate globules. This indicates that silicate minerals were mobile after the carbonate veinlets formed, and similar evidence was presented by other groups at the 28th Lunar and Planetary Science Conference. (3) Valley et al. note that the near-absence of hydrous minerals in ALH 84001, long cited as a problem for a low-temperature origin of the carbonates, is not actually a problem at all. There are many instances on Earth where low-temperature carbonate veins cut silicate rocks without formation of hydrous silicate minerals. The oxygen isotope abundance ratios measured by Valley et al. have been confirmed and extended by two other groups using similar ion microprobe techniques: L. Leshin et al. (1997) and J. Saxton et al. (1997). Although there are still some problems with calibrations and interlaboratory biases, it seems indisputable that the carbonates in ALH 84001 contain relatively heavy oxygen (high d18O) and that they are strongly zoned in oxygen isotope ratios from core to rim. However, the meaning of this zoning is quite disputable. Valley et al. have interpreted the zoning as most consistent with carbonate minerals growing, at low temperature, from a fluid that changed composition over time. Their low temperature is consistent with, but not proof of, microbial life. Leshin et al., on the other hand, interpret the oxygen isotope zoning as forming at higher temperatures in a closed system. Higher temperatures here means 250^ÃC, too high for known Earth bacteria, but a far cry from the 500^Ã-700^ÃC suggested by some other investigators. --------------------------------------------------------------------------------- Jull A. J. T., Eastoe C. J., and Cloudt S. (1997) Isotopic composition of carbonates in the SNC meteorites Allan Hills 84001 and Zagami. J. Geophys. Res., 102, 1663-1669. The authors investigated the sources of the carbon in ALH 84001 (and other martian meteorites), especially using radioactive carbon-14 (14C) as a marker for carbonates that formed on Earth. Radioactive 14C forms continuously in the Earth's atmosphere (and from nuclear bomb tests) and forms only sparingly in space, so the abundance of 14C in the carbonates is a clue to how much they have reacted with carbon from Earth. The authors find that most of the carbonate in ALH 84001 contains 14C, so much 14C that it must have either formed on Earth or traded some of its martian carbon for Earth carbon. The carbon in ALH 84001 with the least 14C is also the richest in the stable carbon isotope 13C, and its 13C abundance is the same as measured for martian carbonates in ALH 84001 and other martian meteorites. This work and Jull et al. (1995) are important for understanding terrestrial contamination in ALH 84001. The authors argue that a great proportion of the carbon and oxygen in the ALH 84001 carbonates originated on Earth, and then diffused into the carbonate mineral grains in the meteorite. This argument, if true, lends plausibility to the idea that the PAHs in ALH 84001 are also terrestrial (Becker et al., 1997). However, Wright et al. (1997) suggest that the amount of 14C found here could also mean only limited contamination by Earth carbon. Goswami J. N., Sinha N., Murty S. V. S., Mohapatra R. K., and Clement C. J. (1997) Nuclear tracks and light noble gases in Allan Hills 84001: Pre-atmospheric size, fall characteristics, cosmic ray exposure duration and formation age. Meteor. Planet Sci., 32, 91-96. As ALH 84001 traveled between Mars and the Earth, it was bombarded by cosmic rays, high-energy particles from the Sun and the galaxy. Interactions of cosmic-ray particles with meteorites leave characteristic signatures like the nuclear tracks produced by cosmic-ray heavy nuclei and trace abundances of the noble elements (e.g., neon and argon) resulting from nuclear interactions of cosmic ray protons with meteoritic matter. Here the authors investigated the evidence for cosmic-ray bombardment in ALH 84001 to understand what happened to this meteoroid after it left Mars and before it landed in Antarctica. They found that ALH 84001 formed approximately 4 billion years ago, and spent approximately 17 million years exposed to cosmic rays; these numbers are consistent with results from many other groups. In addition, the authors here deduce that ALH 84001 was approximately 20 centimeters in diameter before it encountered the Earth, and that ~85% of it burnt up as it passed through the Earth's atmosphere. They also suggest that ALH 84001 did not break up into multiple fragments as it fell through the Earth's atmosphere, and so it is also unlikely that additional fragments of this meteorite exist. There may be calls for the Antarctic Search for Meteorites Program, ANSMET, to return to the Allan Hills area of Antarctica to search for more fragments of ALH 84001 rock. The results in this paper suggest that returning to the Allan Hills for martian meteorites would be no more fruitful than collecting elsewhere in Antarctica. In fact, ANSMET field parties have gathered meteorites from the Allan Hills area many times since their first visit in 1976. In that time, only two martian meteorites have been found in the Allan Hills: ALHA 77005 and ALH 84001. These two meteorites are quite different, and could not be separate fragments from a single meteorite fall. --------------------------------------------------------------------------------- Becker L., Glavin D. P., and Bada J. L. (1997) Polycyclic aromatic hydrocarbons (PAHs) in Antarctic Martian meteorites, carbonaceous chondrites, and polar ice. Geochim. Cosmochim. Acta, 61, 475-481. McKay et al. (1996) discovered that ALH 84001 contains polycyclic aromatic hydrocarbon molecules (PAHs) in moderate abundance, found that these PAHs were distinct from meteoritic and terrestrial PAHs, and found that the PAHs in ALH 84001 were intimately associated with the carbonate minerals that host other possible indications of fossil life. Here, the authors evaluate whether the PAHs in ALH 84001 might be contaminants. To see if the association of PAHs and carbonate minerals in ALH 84001 really suggests that they formed together, the authors put carbonate mineral grains in water samples that contained PAHs--a standard--and a sample of Antarctic ice from the Allan Hills. In both cases, the PAHs in the water attached themselves to the carbonate mineral grains. From this result, the authors infer that the PAHs in ALH 84001 might have become associated with the carbonate minerals without any biologic action. To see if the PAHs in ALH 84001 were actually different from those in other sources, the authors analyzed PAHs in the martian meteorite EETA 79001 (both carbonate minerals and bulk rock), in two carbonaceous chondrite meteorites, and in Antarctic ice from the Allan Hills. The PAHs from these other samples are all similar to those in ALH 84001, especially in having strong signals from the few simplest PAHs (called parent or nonalkylated molecules). The ALH 84001 PAHs are most similar to PAHs in carbonate minerals in the EETA 79001; both meteorites have similar simple PAHs and in similar small amounts of big complex PAHs. The carbonates in EETA 79001 are known to be contaminated with carbon and organic molecules from Earth (Jull et al., 1995; McDonald and Bada, 1995), and so probably contaminated with Earth PAHs. So, Becker et al. conclude that the PAHs in ALH 84001 are probably a mixture of PAHs from Antarctic ice and PAHs from carbonaceous meteorites or interplanetary dust, which could have entered ALH 84001 either on Earth or on Mars. They see no clear evidence in the PAHs for a biological origin on Mars, and suggest that amino acids would be better biomarkers than PAHs. This article is important for characterizing the PAHs from Earth that are likely to collect on meteorites as they sit in Antarctica, and would seem to weaken McKay et al.'s case for traces of martian fossils in ALH 84001. But many questions are not yet answered. 1. The PAHs in ALH 84001 are not merely a mixture of PAHs from CM chondrites and from the Allan Hills ice. The ice contains strong signals from the PAHs naphthalene (mass 128) and coronene (mass 300), while carbonates in ALH 84001 contain neither (their Table 1). Other differences are apparent in the relative strengths of some PAH signals, and in the presence or absence of signals from some less-abundant PAHs. Are these differences artificial, for instance because Becker et al. and McKay et al. used slightly different analytical techniques? Or could the differences be real and significant for the origin of the PAHs? 2. The authors here showed that PAHs in water stick strongly to a calcium carbonate mineral, but is this relevant to ALH 84001? Calcium-rich carbonate minerals are rare in ALH 84001; most of its carbonate is rich in magnesium and iron. Further, the calcium carbonate used in the experiments was not characterized, and may not have the same crystal structure as the carbonates in ALH 84001 (calcite vs. aragonite vs. vaterite structure types); PAHs may bond differently to different carbonate mineral structures. 3. Becker et al. suggest that the PAHs in ALH 84001 are associated with the carbonate minerals because their experiment showed that PAHs in water stick strongly to a carbonate mineral. But do PAHs prefer to stick to carbonates compared to the other minerals ALH 84001? The experiments of Becker et al. shed no light on this question. --------------------------------------------------------------------------------- Bradley J. P., Harvey R. P., and McSween H. Y. Jr. (1996) Magnetite whiskers and platelets in ALH 84001 Martian meteorite: Evidence of vapor phase growth. Geochim. Cosmochim. Acta, 60, 5149-5155. McKay et al. (1996) found that submicroscopic magnetite grains in the ALH 84001 carbonate globules are ". . . cuboid, teardrop, and irregular in shape" and have ". . . no structural defects." These magnetite crystals are similar to crystals produced by bacteria on Earth, and so McKay et al. suggested that the magnetites in ALH 84001 could have been made by martian bacteria. The authors here show that the submicroscopic magnetite grains also occur in other shapes and with structural defects. Using transmission electron microscopy, the authors discovered whisker-shaped magnetite crystals, five times as long as they are wide (10 millionths of a millimeter by 50 millionths of a millimeter). Many of these magnetite whiskers contain a common kind of structural defect, a screw dislocation. The authors also discovered blade- and plate-shaped crystals of magnetite, and many of them contain a structural defect called twinning. On searching through other technical papers, the authors found that magnetite (and similar substances) grow in whisker shapes only from hot gases, hotter than 500^ÃC. Hot gas like this occurs in nature near volcanos, in structures called fumaroles, where the hot gases from a volcano or lava flow escape into the air. In fact, whisker-shaped magnetite crystals were reported from a fumarole deposit in Indonesia by Symonds (1993). Also, Bradley et al. could find no descriptions of bacterial magnetites that were blade shaped, plate shaped, whisker shaped, or that contained structural defects. The authors conclude that the magnetites in the ALH 84001 carbonate globules formed at high temperatures, and not from biological processes. In addition, they note that the magnetite whiskers are approximately the same sizes and shapes as some of the possible fossilized bacteria shown in the McKay et al. (1996) paper. This work can be viewed in two ways: as a refutation of McKay et al.'s claims that the magnetites were made by microorganisms; or as an ambiguous result that merely shows that McKay et al. were a bit exuberant in claiming that all the magnetite crystals were structurally perfect. In the first view, it is clear that some of the magnetite crystals in the ALH 84001 carbonates do not have the shapes and structures of common biogenic magnetites. This fact alone can be seen as a refutation of part of the McKay et al. hypothesis. Because magnetite has a cubic crystal structure, it almost always grows as cubes, octahedra, or other compact shapes. Elongated magnetite crystals are known to grow only from high-temperature gases, whether in nature or in the laboratory. And this inference of high temperature, while not conclusive, is certainly inconsistent with life. In the second view, most of the arguments in Bradley et al. (1996) are ambiguous. While they all are interesting observations, none of them invalidates the hypothesis of McKay et al. 1. From the description in their paper, it is not clear that Bradley's magnetites are from the same layers and veins as the magnetites studied by McKay et al. 2. Although Bradley et al. did find structurally imperfect whisker-shaped magnetites, it would still appear that most of the magnetite crystals in the ALH 84001 carbonates are structurally perfect cuboids (and similar shapes). So far, there is no proof that the whisker and cuboid magnetites formed at the same temperature. 3. To support a high-temperature origin for the ALH 84001 magnetites, Bradley et al. refer to Symonds (1993), who found that whisker-shaped magnetite crystals grew from the hot gases given off by a volcano. But Symonds suggested that temperature alone did not control whether the magnetite crystals grew as cubes or whiskers. In fact, the highest-temperature magnetites grew as cubes, while the whisker-shaped crystals formed at lower temperatures where they grew very quickly (i.e., the gas was very supersaturated). Whisker-shaped magnetites apparently have not been reported in low-temperature carbonate deposits, but it is quite possible that no one has looked carefully. Bradley et al. (1997) will present these results and more at the Lunar and Planetary Science conference this week. Thomas-Keprta et al. (1997) will counter with information that some bacteria do produce elongated magnetite crystals. --------------------------------------------------------------------------------- Shearer C. K., Layne G. D., Papike J. J., and Spilde M. N. (1996) Sulfur isotope systematics in alteration assemblages in martian meteorite ALH 84001. Geochim. Cosmochim. Acta, 60, 2921-2926. The element sulfur has two stable (not radioactive) isotopes, 32S and 34S. The relative abundances of these sulfur isotopes, called the sulfur isotope ratio, can be affected by chemical processes, including metabolism by bacteria. Many Earth bacteria can "eat" sulfur compounds and use them as fuel for growth. Sulfur processed this way by bacteria is typically very depleted in 34S compared to the starting sulfur. Nonbiological processes can enrich or deplete sulfur in 34S, but usually not so much as biological processes. The authors here analyzed the isotopic composition of three pyrite grains associated with the carbonate globules of ALH 84001. The pyrite grains all were enriched 34S compared to the solar system average; in the jargon, they had d34S (pronounced "delta thirty-four S") between +5 and +8 "per mil" (or parts per thousand). These enrichments in 34S suggest that the pyrite (and also the carbonate globules) formed at "low" temperatures, and that the sulfur in the pyrite was probably never processed by bacteria like those on Earth. However, these results are ambiguous because the isotope ratio in Mars' starting sulfur is not known well. If Mars' starting sulfur has d34S near zero (the solar system average), the high d34S of the pyrites could not come from biological processing, at least by bacteria like those on Earth. Nor could the high d34S develop during high-temperature chemical processes. More likely, the pyrites grew from alkaline, oxygen-poor water at less than 150^ÃC. Lunar soils also have a high d34S, which develops as meteorite impacts vaporize some of the soils. On the other hand, if Mars' starting sulfur was rich in 34S (had a high value of d34S), the isotopic composition of the sulfur could be consistent with either a high temperature or a biogenic origin. At high temperatures, inorganic processes do not separate the sulfur isotopes well, so a fluid rich in 34S would deposit pyrite rich in 34S. Acidic waters at low temperature also would not separate sulfur isotopes well, so a fluid rich in 34S would deposit pyrite rich in 34S. If Earth-type sulfur-eating bacteria were fed sulfur that was very rich in 34S, they would accumulate in them sulfur that was not so rich in 34S, perhaps similar to the sulfur in the pyrites. Of course, if martian bacteria process sulfur differently from Earth bacteria, all bets are off. It is easy to think that a low formation temperature for the carbonates in ALH 84001 means that they formed from martian life. But temperature and biology are separate issues. Here, Shearer et al. infer that the pyrite and carbonates in ALH 84001 formed at low temperature without life! Since this work was published, Greenwood et al. (1997) have also analyzed the isotopic composition of sulfur in ALH 84001, and in martian meteorites that have no known or suspected signs of life in them. For ALH 84001, Greenwood et al. got essentially the same sulfur isotope values as this paper; for the other martian meteorites, Greenwood et al. got d34S values between about +3 and -3. These low numbers, so close to the average for the solar system, suggest that Mars' original sulfur was not very different from the solar system average, and so support Shearer's inference of a nonbiological origin for the pyrite. The temperature of pyrite formation is not clear yet: Shearer et al. suggest low temperature, while Greenwood et al. suggest high temperature. This work is continued in Shearer and Papike (1997) and Shearer (1997). --------------------------------------------------------------------------------- * After Science magazine published McKay et al.'s (1996a) article suggesting that they had recognized traces of ancient martian life in ALH 84001, many scientists wrote letters to Science disputing all or part of their results. Science collected these comments and responses to them as "Evaluating the evidence for past life on Mars," Science, 274, pp. 2119-2125. These summaries and commentaries are in the order that Science presented the originals. Anders E. (1996) Science, 274, 2119-2121. After praising the quality and depth of their observations, Anders comments that McKay et al. (1996a) did not consider nonbiological explanations for their discoveries: "For all these observations, an inorganic explanation is at least equally plausible, and, by Occam's Razor, preferable." Anders then suggests nonbiological explanations for most of the chemical evidence for martian life in McKay et al. Anders raises two objections to the description of PAHs in ALH 84001 as implying biogenic activity. First, PAH molecules form as readily from nonbiological chemical compounds as from biological compounds. Given enough time and/or an elevated temperature, PAHs form readily from other organic materials; this process is documented in nature and utilized in industry. Second, the spatial association of PAH molecules and the carbonate globules could have arisen without life. Formation of PAHs can be accelerated (i.e., catalyzed) by the mineral magnetite, and submicroscopic grains of magnetite are abundant in the carbonate globules. Anders also presented five objections to the arguments of McKay et al. concerning the minerals and chemical zoning of the carbonate globules. 1. The chemical zoning patterns in the carbonate globules could be a natural result of mineral solubilities, and need not imply the action of life. 2. The association of magnetite, iron sulfides (pyrrhotite), and carbonate minerals in ALH 84001 could form without the presence of life, as similar associations have formed in the carbonaceous chondrite meteorites. 3. The areas of partially dissolved carbonate minerals could form at normal temperatures and water compositions, without the action of life. 4. The greigite(?) iron sulfide mineral that McKay et al. found was not characterized well, and was not compared with nonbiogenic greigite. Without this comparison, one cannot tell if the greigite(?) is actually relevant to the question of life. 5. Finally, the structure of the carbonate globules (claimed by McKay et al. to be evidence for a biological origin) was not compared to the structures of carbonate globules formed without assistance from life. Without this kind of comparison, one cannot tell if the structures of the carbonate globules are relevant or not. Before the matter of ancient martian life in ALH 84001 is completely resolved, all of Anders' points will need to be studied. McKay et al. (1996b) and Clemett and Zare (1996) provide some answers in their responses to this comment. The fundamental issue behind Anders' comment is scientific proof itself. Can the martian-life-in-ALH-84001 hypothesis be examined piece by piece, one line of evidence at a time? Or must all the evidence be considered together, as one complete package? In natural sciences, it is rarely possible to prove that an idea is true--"proof" consists mostly of showing that an idea fits ("is consistent with") all the facts, and that all other ideas don't fit the facts or are too complicated. Most often, though, scientists can think up many different ideas that can fit all the facts. Then, they will commonly quote "Occam's Razor," which states that the simplest idea is more likely than the complicated ideas. Unfortunately, what is simple to one scientist is needlessly complex to another. McKay et al.'s paper and Anders' comment use different ideas of simplicity, and so arrive at different preferred conclusions. McKay et al. invoked Occam's Razor (without naming it) in justifying a biological origin for all their observations: "Although there are alternative explanations for each of these phenomena taken individually, when they are considered collectively, particularly in view of their spatial association, we conclude that they are evidence for primitive life on early Mars." From this perspective, McKay et al. did not need to consider nonbiological explanations for each observation, only nonbiological explanations of the all of the observations at once. They did not find any nonbiological explanations, and so had to accept the idea of martian life. On the other hand, Anders invoked Occam's Razor (quoted above) to justify nonbiological processes for each individual observation. Anders did not search for a single nonbiological explanation for all the evidence, and did not consider how likely it was that all of his proposed processes could have affected small areas in a single rock. To some extent, then, Anders and McKay et al. are not looking at the evidence in the same way; McKay et al. are "holists," and Anders is a "reductionist." For the possible martian fossils, it remains to be seen which view of the world* is more useful. * "Weltanschauung" to the philosophers. --------------------------------------------------------------------------------- Shearer C. K. and Papike J. J. (1996) Science, 274, 2121. Here, the authors summarize their sulfur isotope measurements that were reported earlier in Shearer et al. (1996), which are described below. Shearer and Papike emphasize that the pyrite mineral grains that they analyzed earlier are related to the carbonate globules, and that the sulfur in the pyrite is enriched in the stable isotope 34S compared to the solar system average. Sulfur-eating bacteria on Earth produce mineral-like pyrite that is strongly depleted in 34S, so it is unlikely that the pyrite in ALH 84001 was made by Earth-type bacteria. Martian bacteria could still be involved, however, if Mars itself was much richer in 34S than the Earth is, or if martian bacteria process sulfur differently from Earth bacteria. For more detail, see the discussion of Shearer et al. (1996) below. Gibson et al. (1996) respond directly to this comment. McKay et al. (1996a) did not claim that the pyrite in ALH 84001 was biogenic, so, strictly speaking, this report by Shearer and Papike is not relevant to the current hypothesis of ancient martian fossils in ALH 84001. However, the pyrite crystals are spatially associated with the carbonate globules, and it would have seemed reasonable that the pyrite and the carbonates grew from the same fluids with the same sulfur isotope abundances. On the other hand, if the pyrite had a deficiency of 34S (such as might be expected from biogenic pyrite on Earth), it might possibly have been cited by Gibson et al. (1996) as further evidence of biogenic activity in ALH 84001. This work has continued in Shearer (1997) and Shearer and Papike (1997). --------------------------------------------------------------------------------- Bell J. F. (1996) Science, 274, 2121-2122. Bell's comment centers on the PAH organic molecules found in ALH 84001 by McKay et al. (1996); Bell accepts that these PAHs are martian, but not that they imply martian life. He suggests that the PAHs may have come from meteorites falling onto Mars, just as a few percent of the Moon's soil is made of meteorite debris. Specifically, Bell suggests that the PAHs in ALH 84001 came from material like the C2 carbonaceous chondrite meteorites, and suggests that the sources of this C2 material included the moons of Mars, Phobos and Deimos. McKay et al. (1996a) and Becker et al. (1997) agree with Bell that the PAHs in ALH 84001 are similar to those in the C2 carbonaceous chondrites. The PAHs in these meteorites are not identical, but are they similar enough to suggest a common origin? Bell and Becker say "yes," McKay et al. say "no, especially in light of the associated evidence." Bell is correct that a few percent of the lunar soil is made of meteoritic material like C2 carbonaceous chondrites (a point I mistakenly disputed in earlier versions of this commentary). Although few meteorites are carbonaceous, the vast majority of interplanetary dust is like C2 carbonaceous chondrites, and that dust makes up most of the mass that falls onto planets. The moons of Mars are very dark; their darkness might be from the carbon in carbonaceous chondrite material, but their darkness might have other causes (Murchie et al., 1991). --------------------------------------------------------------------------------- Clemett S. J. and Zare R. N. (1996) Science, 274, p. 2122-2123. Clemett and Zare are among the authors in the original McKay et al. paper, and they respond to comments of Anders and Bell related to PAHs, the organic molecules called polycyclic aromatic hydrocarbons. Clemett and Zare emphasize that the PAHs they found in ALH 84001 are not laboratory contaminants, and are apparently only a small part of all the organic materials in ALH 84001. They agree with Anders (1996) that some of the PAHs in ALH 84001, the low-mass ones, could have formed by inorganic processes at high temperature. The high-mass PAHs, although less abundant, are very similar to the break-down products of kerogen, a variety of solid organic material that is common on Earth and in the carbonaceous chondrite meteorites. Earth kerogen formed from living matter, and meteorite kerogen did not. Clemett and Zare leave with two questions: how could nonbiologic kerogen get into an igneous rock, one that solidified from molten lava; and how could nonbiologic kerogens (or PAHs) come to be associated only with the carbonate globules in ALH 84001? As an aside, Clemett and Zare also reply to comments from Simoneit and Hites and from Requejo and Sassen, but neither of these comments was printed in Science. Clemett and Zare agree that some nonbiological processes could have produced the distribution and abundances of PAHs that they observed in ALH 84001: the low-mass PAHs could be the product of inorganic reaction at high temperature, and the high-mass PAHs could form by the low-temperature reaction of inorganic kerogen. But the issue is whether the PAHs in ALH 84001, in their association with the carbonate globules, are more easily explained by biological or nonbiological mechanisms. A nonbiological scenario would have to start with carbon-rich gas reacting at high temperature to form the low-mass PAHs. Then, nonbiologic kerogen, from some other source, would have to decompose at low temperature into high-mass PAHs. Either the kerogen or the high-mass PAHs would have to infiltrate ALH 84001, adhere only to the carbonates, and not displace the low-mass PAHs already in place. Is this sequence of events actually simpler and more believable than the growth, death, and decomposition of martian bacteria? --------------------------------------------------------------------------------- McKay D. S., Thomas-Keprta K. L., Romanek C. S., Gibson E. K. Jr., and Vali H. (1996b) Science, 274, 2123-2125. Here, McKay et al. respond directly to Anders' (1996) comments about minerals in the carbonate globules and about the morphology of possible fossil shapes in ALH 84001; Clemett and Zare responded to Anders' comments on PAHs. Anders' comments stressed the similarity of the carbonate globules and their minerals to some grains in the CI carbonaceous chondrite meteorites. McKay et al. agree that similarities are present, but emphasize the significant differences between ALH 84001 and the CI carbonaceous chondrites. In particular, ALH 84001 is an igneous rock, while the CIs have been altered at low temperatures to clays, serpentine, and similar water-bearing silicate minerals. McKay's responses to Anders' comments are keyed to Anders' points (as above). 1. McKay et al. agree with Anders that the chemical zoning pattern in the carbonate globules could have been produced by inorganic crystallization. They stress, however, that the repetitive (oscillatory) zoning pattern and composition difference between one globule and another can only arise from complex inorganic processes. 2. Anders compared the carbonate-magnetite-sulfide minerals in ALH 84001 to those in CI carbonaceous chondrite meteorites. McKay et al. respond that, in effect, the CIs are not good analogies. Magnetite grains in carbonate minerals are much larger in CIs than in ALH 84001. And magnetite grains in carbonate minerals in CIs do not have cuboid shapes as they do in ALH 84001. 3. McKay et al. agree with Anders that the partially dissolved carbonate grains in the carbonate globules could have formed in nearly neutral (nonacidic or alkaline) water, and do not require the moderate acidity invoked in McKay et al. (1996a). McKay et al. restate that the globular morphology of the ALH 84001 carbonates is similar to those formed by bacteria on Earth, and unlike the carbonate areas formed inorganically in the CI carbonaceous chondrites. They stress, however, that no matter what the exact water composition, no simple inorganic process can form all the observed structures and minerals in the carbonate ellipsoids. 4. On the matter of greigite(?) in ALH 84001, Anders had hoped to see it compared to nonbiogenic greigite. McKay et al. respond that life seemed to be involved with the formation of all greigite on Earth, at least all the greigite that they were aware of. Living organisms either produce greigite directly themselves, or produce the hydrogen sulfide gas that goes to form greigite. 5. Anders commented that the structures of the carbonate globules should have been compared directly to carbonates that grew without assistance from life. McKay et al. respond that the shape of possible fossil forms is not yet definitive proof that they are real fossils, that similar shapes have not been found in lunar or asteroidal meteorite samples, and that more work is needed. They also agree with Anders that a proof that the fossil shapes actually are fossils would make all the other arguments irrelevant. I see two underlying themes in this response: that ALH 84001 is unique, and that the minerals and structures of the carbonate globules are too complex for any simple inorganic processes. There is, of course, no doubt that ALH 84001 is unique. But Anders and McKay et al. disagree on whether the carbonate globules in ALH 84001 are so unusual that seemingly similar structures in the CI carbonaceous chondrites are not relevant. It has been suggested that the CI carbonaceous chondrites are from Mars (Brandenburg, 1996), but most evidence seems to suggest otherwise (Treiman, 1996). McKay et al. emphasize the complexity of the carbonate globules, both in the chemical zoning of their carbonate minerals and in the groupings of minor minerals in the carbonates. The complexity alone suggests to them the action of complex biological systems, and they want to consider all the evidence in McKay et al. (1996a) as a systematic whole, and not as a set of separate pieces. Quoting from their response, the formation of the carbonate globules ". . . cannot be simple equilibrium . . . , and must include changing conditions and kinetic effects. Whether such models are more plausible than biogenic models is a matter of judgment." As an aside, the fifth point of Anders' comments seems to refer to the shapes and structures of the carbonate globules, while McKay et al. here responded about the sausage-shaped things that might be fossil bacteria. Some critical sentence or idea may have been lost. --------------------------------------------------------------------------------- Gibson E. K. Jr., McKay D. S., Thomas-Keprta K. L., and Romanek C. S. (1996) Science, 274, 2125. The authors respond directly to Shearer and Papike's (1996) claim that sulfur isotope ratios on pyrites near the carbonate globules probably mean that they formed without help from bacteria. The authors note that the pyrite grains may not be relevant to McKay et al.'s hypothesis because pyrite is not in the carbonate globules, it did not grow with the structurally flawless submicroscopic magnetites, and is not associated with the PAHs. The submicroscopic sulfur-bearing minerals, pyrrhotite and greigite, which are part of McKay et al.'s hypothesis, will be very difficult to analyze for sulfur isotope ratios. These sulfur-bearing grains are so small that the carbonate and magnetite grains around them would also end up being analyzed for sulfur. The carbonate and magnetite grains don't contain sulfur, but they do contain lots of oxygen, and oxygen molecules can masquerade as sulfur atoms in these isotope analyses. Sulfur atoms with mass 32, 32S, can be mimicked by the oxygen molecule 16O16O; and sulfur atoms with mass 34, 34S, can be mimicked by the oxygen molecule 16O18O. Greenwood et al. (1997) report that the isotope ratio for sulfur from the carbonate globules, presumably from pyrrhotite, is nearly the same as for the pyrite grains. Their sulfur isotope of the pyrrhotite analyses are quite imprecise (d34S somewhere between +12 and -1), and it is not clear if they considered the possible interferences from molecular oxygen. --------------------------------------------------------------------------------- References Bogard D.D. and Garrison D.H. (1997) 39Ar-40Ar age of ALH 84001. In Clifford S. and Treiman A.H. eds., Conference on Early Mars: Geologic and Hydrologic Evolution, Physical and Chemical Environments, and the Implications for Life. L.P.I. Contribution 916, 10-12. Bradley J. P., Harvey R. P., and McSween H. Y. Jr. (1997) Magnetite whiskers and platelets in the ALH84001 martian meteorite: Evidence of vapor phase growth. Lunar Planet. Sci. XXVIII. Brandenburg J. E. (1996) Mars as the parent body of the CI carbonaceous chondrites. Geophys. Res. Lett., 23, 961-964. Greenwood J. P., Riciputi L. R., and McSween H. Y. Jr. (1997) Sulfur isotopic variations in sulfides from shergottites and ALH 84001 determined by ion microprobe: No evidence for life on Mars. Lunar Planet. Sci. XXVIII. Jakosky B.M. and Jones J.H. (1997) The history of Martian volatiles. Rev. Geophys. 35, 1-16 Jull A. J. T., Eastoe C. J., Xue S., and Herzog G. F. (1995) Isotopic composition of carbonates in the SNC meteorites Allan Hills 84001 and Nakhla. Meteoritics, 30, 311-318. Kjarsgaard B.A. and Hamilton D.L. (1989) The genesis of carbonatites by immiscibility. 388-404 in Bell ed. Carbonatites: Genesis and Evolution, Unwin-Hyman, Boston. Leshin L., McKeegan K. D., and Harvey R. P. (1997) Oxygen isotopic constraints on the genesis of carbonates from martian meteorite ALH 84001. Lunar Planet. Sci. XXVIIII,805-806 McDonald G. D. and Bada J. L. (1995) A search for endogenous amino acids in the Martian meteorite EETA79001. Geochim. Cosmochim. Acta, 59, 1179-1184. McKay D. S., Gibson E. K. Jr., Thomas-Keprta K. L., Vali H. , Romanek C. S., Clemett S. J., Chillier X. D. F., Maechling C. R., and Zare R. N. (1996a) Search for past life on Mars: Possible relic biogenic activity in martian meteorite ALH 84001. Science, 273, 924-930. Murchie S., Britt D., Head J., Pratt S., Fisher P., Zhukov B., Kuzmin A., Ksanfomality L., Zharkov A., Nikitin G., Fanale F., Blaney D., Robinson M., and Bell J. (1991) Color heterogeneity of the surface of Phobos: Relationships of geologic features and comparison to meteorite analogs. J. Geophys. Res., 96, 5925-2945. Saxton J. M., Lyon I. C., and Turner G. (1997) Oxygen isotope ratio zoning in ALH 84001 carbonates (abstract). In Conference on Early Mars: Geologic and Hydrologic Evolution, Physical and Chemical Environments, and the Implications for Life. (S. Clifford et al., eds.), pp. 70-72. LPI Contribution No. 916. [http://cass.jsc.nasa.gov/lpi/meetings/earlymars/pdf/program.pdf] Shearer C. K. (1997) Sulfur isotopic systematics in ALH 84001. Open-and closed-system behavior of sulfur in a martian hydrothermal system. Lunar Planet. Sci. XXVIII. Shearer C. and Papike J. (1997) The petrogenetic relationship between carbonates and pyrite in martian meteorite ALH 84001. Lunar Planet. Sci. XXVIII. Symonds R. (1993) Scanning electron microscope observations of sublimates from Merapi Volcano, Indonesia. Geochem. J., 26, 337-350. Thomas-Keprta K. L., Romanek C. S., Wentworth S. J., McKay D. S., Fisler D., Golden D. C., and Gibson E. K. (1997) TEM analysis of fine-grained minerals in the carbonate globules of martian meteorite ALH 84001. Lunar. Planet Sci. XXVIII. Treiman A. H. (1996) Comment on "Mars as the parent body of the CI carbonaceous chondrites" by J. E. Brandenburg. Geophys. Res. Lett., 23, 3275-3476. Valley J.W., Eiler J.M., Graham C.M., Gibson E.K.Jr., Romanek C.S., and Stolper E.M. (1997) Low-temperature carbonate concretions in the martian meteorites ALH 84001: Evidence from stable isotopes and mineralogy. Science 275, 1633-1638. Wadhwa M. and Lugmair G. W. (1996) The formation age of carbonates in ALH 84001 (abstract). Meteoritics 31. A145. Wright I. P., Grady M. M., and Pillinger C. T. (1997) Evidence relevant to the life on Mars debate. (1) 14C results. Lunar Planet. Sci. XXVIII.