1. Introduction

New collections of interplanetary dust particles (IDPs) are allowing identification of early and pre-solar system material, including prebiotic organics and potential bio-fossils. These IDPs originate from comets and fragmenting collisions of asteroids, moons and planets, or may include dust grains from the pre-solar nebula.

Though IDPs and meteorites are largely composed of mineral materials, they also comprise substantial carbonaceous components. Carbonaceous chondrite (CC) meteorites have 10% or more carbon, while cometary dust has a 30% CHON particle fraction (made of the light elements C,H,O,N) as well as mineral particles (“ROCK”) and mixed particles, as discovered by the comet Halley space-probes. IDPs recovered from the stratosphere are agglomerates that could be purely meteoritic or could have been processed in comets. Isotopically anomalous sub-micron components show the inclusion of pre-solar grains, while the silicate components of ROCK indicate material condensed during an energetic phase of the early sun.

Traces of water can be detected in some samples and frequently the minerals in CC meteorites show evidence of aqueous alteration. The identification of clay particles ejected from comet Tempel-1 by the Deep Impact probe in 2001 implied that aqueous alteration may indeed proceed on comets. The early IDP studies in the 1970s saw many of these particles as coming from comets; meanwhile, evidence for CC meteorites also originating from comets has been increasing. Enhanced IDP collection via balloon-borne cryosampling (Lal et al., 1996) and use of analytic techniques down to sub-micron scales have in recent years opened up further progress in studying fossils in both the mineral and the carbonaceous components.

Recovery of high velocity IDPs from the stratospheric is effective because the Earth's tenuous air at ~100 km gradually slows down particles under 100μm while the smaller ones (<20μm) are only moderately heated (depending on density and zenith angle; Coulson and Wickramasinghe 2003), so lose their volatiles but retain more complex organics. They slow up to low terminal velocities (some cm/s) and take weeks to months to descend below the stratosphere.

In the 1970s, high altitude flights (U2 aircraft) were used to collect them from the lower stratosphere, 18-20km altitude, on oiled sheets exposed outside aircraft flying at ~200m/s. This method suffered from contamination as well as breakage of the particles and bias against small light ones (which tend to divert with the air stream).

The resultant bank of IDPs, held by NASA, has given us the basic data for IDP systematics.

The Indian Space Research Organisation (ISRO) balloon flights from the 1990s initially reached ~30 km for smapling stratosphere CFCs, collecting frozen air in steel cylinders with all-metal valves (remotely controlled) immersed in liquid neon. The more recent flights reach 40-45 km with all equipment ultra-clean and sterile to reduce contamination (Lal et al., 1996); terrestrial particles are rare and spacecraft debris micro-particles are relatively easily spotted.

These particles were collected in the following manner: one set of samples is extracted from the cylinders by issuing the compressed air through micropore filters; another set is from filtering washings of particles adhering to the interior surface. At Cardiff we have used a few glass fibre filters, but mainly 0.45μm acetate filters. For better identifying the carbon fraction, some samples were transferred to silicon wafers (Miyake 2009). The rate of particle recovery is several times higher than the old NASA collections.

2. Early Claims of Fossils in Meteorites

Early in the 1960's, Claus and Nagy (1961) identified possible microfossils in CCs, supported by bio-indicators. Critics mounted a wide array of objections to discredit these discoveries, claiming, for example, that some fossils had been contaminated by pollen. Recent studies established, however, that forms called “organised elements” by Nagy et al. (1962) are readily found in freshly fractured surfaces of CC meteorites (Hoover 2006a) and can no longer be dismissed as pollen contaminants. Timofejew (1963) identified near spherical microscopic objects and a subsequent review (Manten 1966) described a range of smaller (4-30μm) spherical structures showing double cell walls, spines, furrows, pores and flagella-like forms.

Methods for isolating fossil carbonaceous material from sedimentary rocks - dissolving the minerals with acids - were applied to the Orgueil meteorite by Rossignol-Strick and Barghoorn (1971). They discovered hollow spherical shells, which in terrestrial rocks would have presumed bio-origin, but, according to the authors, could result from carbonaceous deposits on mineral particles.

That debate over non-biotic artefacts resembling bio-fossils has persisted, and a range of bio-indicators has been called up in evidence favouring biotic origins (Hoover 2006a). Likewise, claims of bio-fossils in the martian meteorite ALH84001 have been explained away as contamination or as non-biotic artefacts. However, McKay et al. (2009) recently reported studies of potential biofossils (carbon-associated "biomorphs‟) in additional martian meteorites Nakhla and Yamato-593.

Chemical studies have shown meteoritic carbonaceous material to be highly complex (analogous to kerogen or sporopollenin; Brooks and Shaw, 1969). A range of extractable organic compounds, including alkanes, alkenes, aminoacids and nitrogen heterocyclics, extending to the purine and pyrimidine bases of DNA, have been reported (Hayatsu and Anders 1981). However, these were presumed to derive abiotically from dust and condensing gases processed in interstellar space by stellar UV and other radiation.

Seeing that such abiotic carbonaceous material would reach the early Earth along with the meteorites, Deamer (1985) was motivated to extract non-polar components from the Murchison CC with chloroform/methanol. On drying and rehydrating the extract with alkaline phosphate buffer, he obtained membrane-like structures and suggested this source for self-organising pre-biotic chemicals necessary for generating elementary living forms. Why CCs contain these particular molecular pre-biotics naturally remained puzzling.

The Cronin et al. review (1988) said the 1960/70s argument over biogenic or abiogenic origin had been resolved in favour of the latter, though they judged no particular mechanism was adequate to explain how such abiotic matter was created (“multiple processes and/or environments”). However, recent studies indicate otherwise. Mukhopadhyay et al. (2009) believe the complex hydrocarbons may derive from bacteria and/or primitive algal remains (based on SEM-EDS, visual kerogen analysis using fluorescence, and white-light microscopy). Those authors found abundant alkanes (normal, cyclo-, and isoalkanes), alkyl aromatics, some polycyclic aromatic hydrocarbons, thiophenes, and nitrile compounds with biological signatures, especially within the Tagish Lake and Orgueil meteorites.

Microfossils have in recent years been found in every CC meteorite studied by Hoover and co-workers in Russia (Moscow, Paleontological Inst) and USA (MSFC), but notably not in non-CC meteorites (Hoover 2006a,b). They have used knowledge of microorganism morphology to tentatively identify some bio-fossils in CCs. Their ESEM and FESEM images of artifacts in the Murchison and Orgueil carbonaceous meteorites have uncovered filaments that typically exhibit dramatic chemical differentiation between the putative microfossil and the adjacent meteorite matrix.

The first suggestion of bio-fossils in IDPs was made by Hoyle et al. (1985) from an image of filamentous object (0.1x1.5μm) studied by Bradley et al. (1984). They identified a carbonaceous tubule containing magnetite crystals, similar to terrestrial fossils of iron oxidising bacteria.

3. Identification of Acritarchs

Organic-walled microfossils found in terrestrial sedimentary rocks but of unidentified species are termed acritarchs (formerly 'hystrichospheres'). Many of the specimens found in association with IDPs, especially the ovoids, clearly resemble acritarchs.

Rossignol-Strick et al. (2005) reviewed the 1971 discovery of the acid-resistant, organic "hollow spheres" by Rossignol-Strick and Barghoorn (1971) and sought new examples in the Orgueil meteorite. Ovoid bodies in their new images were found to be composed of Fe-mineral within a thin carbonaceous sheath, like those tentatively identified as magnetite with ~0.2μm organic coatings (Alpern and Benkheiri 1973). By contrast, the organic globules found in the Tagish Lake meteorite (Nakamura et al., 2002) are mainly small structures - these μm-sized solid ovoids are quite distinct from the 1971 acid-resistant hollow spheres discovered by Rossignol-Strick and Barghoorn. The 5μm coccoid with 0.3μm carbonaceous envelope reported by Hoover (2006b) plus a similar one which Mukhopadhyay et al. (2009) mapped in carbon and sulphur do, on the other hand, correspond to the 1971 discovery of acritarch-like structures in the Orgueil.

The Cardiff collection of IDPs contains many more acritarch examples. Those shown in Figs. 1-3 were found by the first SEM studies (Wallis et al., 2002). The single spheres are spore-like, sometimes damaged (Fig. 1E) and often showing cracks (Figs.1 B, C, D, F). Cracks in the 'spores' are sometimes seen to widen under the microscope, but breaks in the surface appear to have existed pre-preparation. The surfaces show diverse structure and coatings - Fig. 1C shows partial coating, while the examples of Fig. 2 show thicker mineral deposits. Fibres (straight rods) about 0.5μm diameter are commonly attached (eg. Figs.2 A, C, E, F), while D shows finer whiskers embedded in the coating (Wallis et al., 2006).

Acritarchs-like structures were also discovered in stratospheric dust studied by Miyake (2009). Fig.4 shows cases of 9-10μm carbonaceous spheres, with partial mineral coatings that have cracked under an apparently uncracked shell (A, E). The EDX spectrum D for particle C shows the shell is high in C (~ 70%) with coating of Na, Cl and O (minor amounts of K, Ca, Si). Such mineral encrustations show they are not 'spores', nor do the surface structures resemble pollen. The specimen E appears to have fossilised flagella while F has fine whiskers in the thicker surface deposit. A, B and C show electron-transparent sheaths (arrowed white) suggestive of extracellular polymer substances (termed EPS), resembling those that Hoover (2006b) has reported from meteorite studies.

Fig.5 shows an acritarch with fine whiskers both protruding and embedded in the surface. In this case the SEM heating is seen to have cracked the gold coating, as shown by the absence of gold in the spectrum S2 taken on the crack. Fig.6 shows EDX studies of acritarchs on a silicon base, which avoids confusion of the acetate filter carbon with particle carbon. Figs. 6a and 6b both show high carbon, but only the second has a mineral coating, 6a having a sheath with little Na and Cl, no Ca but of structureless carbonaceous material, indicative of EPS. Fig. 6c indicates complex and inhomogeneous coating processes.

Fig.7 shows another variety of acritarch, smaller and with distinctive surface structure. The C is high (58%), O low (4.3%) and P undetected, so it is not a terrestrial pollen contaminant. The cracking under SEM heating again indicates hollow spheres.

4. Siliceous Fibres Abundant in IDPs

Such fibres are common in recovered IDPs as isolated rods or attached to other grains. Initially we assumed these to be terrestrial contaminants (glass fibres). However, fibres are both attached or associated with acritarchs, as seen above, and embedded in loose aggregates, as shown in Fig.8.

Sample A shows a mineral silicate particle with significant metals (Al, Fe etc.), partially coated with NaCl which also coats the attached fibres. C,D,E are looser (more porous) aggregates (the spectrum F shows a mineral silicate) thought to be formed in comets, with examples of embedded and attached fibres. D has both carbonaceous and magnetite components as well as the sub-micron siliceous whiskers and fibre fragments.

The single fibres in Fig. 9 are over 10μm long, and down to 0.4μm diameter. We term these 'fibres' (near whole exoskeletons of tubular specimens) and describe the finer ones (0.1μm diameter) as ' whiskers'. C/D shows the largest one found - a 20x3μm cylinder, with diverse submicron rods or whiskers attached to its surface (Wallis et al., 2006).

The spectrum F for sample E shows Si and C with the usual Na-Cl of the coating (crusting at the top). The cylinders are potentially hollow judged by the left end of C and other angular ends (broken as B and E) and thought to be diatoms whose siliceous cell walls include carbon (Miyake et al., 2009). The finer 'whiskers' that are largely embedded in surface coats (eg. Figs.4 E,F) would be the siliceous spines that some diatom species possess in abundance (Miyake et al., 2009). Complexes of siliceous fibres have been found as in Fig.10 - though initially disregarded as terrestrial contaminants, the embedded salt crystal and surface condensates in D indicate a probable cometary origin.

5. Conclusions

We have discounted a possible astrophysical origin of the siliceous fibres in outflows from the sun and stars (Miyake et al., 2009). Asteroidal and cometary silicates are understood as crystalline or fine phyllosilicate clays identified by infra-red emissions in comet Tempel-1's dust ejected by the Deep Impact collision (Lisse et al., 2006). The siliceous rods and fibres fit with neither origin. The proposal of comets as a potential habitat for siliceous diatoms dates to 1985 (Hoover et al., 1986) because of IR spectral similarities and because polar diatoms survive in polar ice at low light levels, hibernating at low temperatures.

Some marine diatoms possess whiskers (pili = hair-like extensions), others have intricate siliceous exoskeletons. We therefore hypothesised (Miyake et al., 2009) that our siliceous fibres and whiskers are fragments of fossil diatoms that lived on comets and/or icy satellites. If embedded in ice ejected via collisions, they would be freed as the ice sublimates away. If on the surface of a comet (or in martian seasonal ice deposits) the fibres would attach to other minerals as the ice sublimates away, becoming embedded in the loose aggregates.

The frequent occurrence of siliceous fibres and whiskers may be a useful indicator of cometary particles (though not all comet particles, eg. the highly porous smoke-like condensates). The association with low density aggregates (Fig. 12C-E) confirms their cometary origin. Their association with acritarchs (Figs 2, 4) links those carbonaceous shells to comets too. As such fibres are not evident or at least not common in CCs, either the identification of CCs with comets was mistaken or the siliceous fibres must tend to disappear in the conversion to more compact CC material. We remark that Rossignol-Strick and Barghoorn (1971) did find siliceous frustules (cell wall fragments) in their Orgueil samples.

Acritarchs when found in terrestrial sedimentary rocks are presumed to be biological, eukaryotes, but of undetermined species. The discovery in the Orgueil meteorite was explained away as mineral artifacts. Our IDP acritarchs would challenge such abiotic explanations. Though both may derive from comets, those aggregated in carbonaceous chondrites have undergone some compaction and aqueous alteration, while the IDP examples may have been released within icy matrices. The IDP samples are less altered geochemically than Orgueil's; they have mineral coatings and are just loosely associated with other materials considered to be of cometary origin. Thus, if C-isotope studies (via eg. an ion microprobe) confirm their space origin, the specimens of Fig. 4 provide a strong indicator of extraterrestrial and probably cometary life.

Terrestrial acritarchs have been recovered from sediments deposited as long as 3.2 Gyr ago, but at about 1 Gyr ago they started to increase in abundance, diversity, size, complexity of shape, and 650 Myr ago their sizes and number of spines increased, the latter presumed to be protective against predation. In our IDPs, we found the second type (Fig.7), but small and with primitive spines. We also found a third, toroidal type (Fig.3 E/F); this unusual type has also been found of 10μm-size in the Tagish Lake meteorite (Rauf et al., 2010a, Fig.1c) and the same shape of 1μm size in the martian Nakhla meteorite (McKay 2009).

The limited diversity in acritarchs in the IDP samples supports arguments for a non-terrestrial origin. They would originate on solar system bodies where darwinian evolution of cell walls protective against predators has hardly progressed. The alternative origin in material blasted off the Earth by giant impacts in geological history, such as formed the 65 Myr Chicxulub crater, would have resulted in more diverse and less primitive forms.

We have found certain other putative biofossils in the balloon-collected IDPs, analogous to some of those studied by Hoover and colleagues in the CC meteorites (Hoover 2006a,b). We find several examples of their 'electron-transparent sheath' (Figs.4 A-C, 6A), amorphous semi-transparent material attached to the surface, known as extracellular polymer substance (EPS). Ice-living bacteria and diatoms produce EPS (Mock and Thomas 2005) to enable metabolic activity at low temperatures (even to -35°C) and prevent damage by ice crystals. We presume that other eukaryotic ice-living protists - including acritarch progenitors - likewise produce EPS, this being vital for survival in cometary ice. Evidence of EPS-like biofilm associated with ~1μm 'fossil microbes' in the Nakhla meteorite is shown in a NASA SEM (McKay 2009).

In conclusion, we do not believe these specimens are of terrestrial origin - ejected from the surface of Earth and returned to the stratosphere after millions of years circulating in the solar system as interplanetary dust. Rather, the IDPs provide a new source of evidence of fossil biology outside the Earth, complementary to that from carbonaceous chondrites. Transfer of both techniques and identifications to and from the ongoing studies of biofossils in martian meteorites (McKay at al., 2009) should be fruitful. Further studies of particles collected by ISRO's stratospheric cryogenic samplers will undoubtedly prove rewarding.

Acknowledgements: The stratospheric air samples were provided by Prof. Jayant Narlikar (IUCAA and leader of the balloon-borne cryogenic mission); Prof. Chandra Wickramasinghe (CCAB) led the Cardiff studies and Dr Anthony Hann provided invaluable help in SEM investigations at Cardiff University.