Since Viking, Mars exobiology has shifted focus to include the search for evidence of an ancient biosphere. The failure of the Viking biology experiments to detect life at the two sites sampled has been broadly accepted as evidence that life is probably absent in surface environments, a view consistent with the lack of liquid water, the high UV flux, and oxidizing conditions observed in the surface soils sampled. The detailed hydrological history of Mars is unknown. But in broad outline, it appears that clement conditions for life probably existed at the surface of Mars early in its history, particularly during the time that widespread valley networks were formed within the ancient cratered highland terranes of the southern hemisphere (Carr 1996). Crater ages suggest that this period of surface water occurred toward the end of late bombardment, perhaps 3.5-4.0 billion years ago, during the time that life was just emerging on the Earth (McKay and Stoker 1989).

Despite the fact that present surface conditions are inhospitable to life, it is quite plausible that life may yet persist in subsurface environments on Mars where liquid water could be present owing to higher temperatures and pressures (Boston et al 1992). However, such environments are unlikely to be explored prior to manned missions, perhaps decades hence. Thus, if life ever arose on Mars, we will likely discover evidence of its former presence in the surface rock record long before we are able to drill for liquid water located deep (perhaps hundreds to thousands of meters) beneath the surface. Clearly, the exploration for a fossil record of life on Mars requires a much different strategy than the search for extant life (see Kerridge et al. 1995), and this strategy is presently embodied in Mars exopaleontology, a new subdiscipline of geology that borrows its scientific heritage from Precambrian paleontology, microbial ecology, biosedimentology, biogeochemistry and Mars surface science (Farmer 1995).

The strategy for Mars exopaleontology is founded on a few basic principles gleaned from studies of the Precambrian fossil record, as well as studies of fossilization processes in modern environments on Earth which are regarded as being good analogs for the early Earth and Mars. Such studies reveal not only the ways in which biological information is captured and preserved in sediments, but also suggest the optimal methods for extracting biological information from ancient rocks on Earth, or from samples returned to Earth from Mars.

It is noteworthy that even if life never developed on Mars, the prebiotic organic chemical record preserved there is an equally important scientific objective for exopaleontology. The absence of a plate tectonic cycle on Mars suggests that old geologic terranes may be much better represented there. The destructive processes of burial metamorphism are also likely to be much less a problem on Mars, although impact-related metamorphism and brecciation of the surface materials is likely to have overprinted the information of the early rock record to some extent. However, the prebiotic chemical record, of vital importance in understanding the origin of terrestrial life, is still likely to be much better preserved on Mars than on Earth. Ancient Martian terranes may provide access to a record of prebiotic processes long ago destroyed on our own planet by plate tectonic and hydrological cycling.

On Earth, >98% of all the organic carbon fixed by organisms is destroyed and recycled. The small amount of organic carbon that escapes degradation persists in the crust only by virtue of having been rapidly buried in fine-grained, low permeability sediments, which isolates organic materials from destructive biochemical processes. This organic carbon reservoir, makes up the chemical portion of the fossil record in which biological information is preserved as various organic biomarker compounds (e.g. hopanes, the degradation products of cell wall lipids), isotopic signatures (e.g. characteristic carbon isotope ratios reflecting biological fractionation processes), and biominerals (ie., those whose precipitation is mediated by biological processes, e.g. phosphates).

The preservation of organic carbon occurs under a very restricted set of geologic environments and conditions which are fairly well understood on the Earth. But even where organic compounds are destroyed by oxidation, biosignatures can still persist in sedimentary rocks as biofabrics produced by microorganisms (e.g. mesoscopic features such as stromatolites or related biolaminated sediments, as well as the characteristic microfabrics contained therein).

A basic tenet that has emerged from paleontological studies of the Precambrian is that the long-term preservation of biological information as fossils is only favored in environments where aqueous minerals precipitate rapidly from aqueous solutions, or where fine-grained, clay-rich detrital sediments accumulate very rapidly, entombing living organisms or their byproducts, before they can be degraded (Farmer 1995). The most favorable aqueous minerals are those that form fine-grained, impermeable host rocks (favoring a closed chemical system) which can isolate organic materials from oxidation. In addition, favorable host minerals are those that are chemically and physically stable and resistant to major fabric reorganizations during diagenesis.

Stated another way, the most favorable host minerals for the long-term preservation of organic materials are those with long crustal residence times. These tend to be minerals that are most resistant to chemical weathering. High priority minerals in this category include silica, phosphates, clays, Fe-oxides and carbonates. These mineral groups are relatively stable chemically, and are commonly precipitated in near surface environments as chemical sediments, or authigenic (early diagenetic) cements. Not coincidentally, such compounds are the most common host minerals for the microbial fossil record on Earth. All of the classic microbiotas of the Precambrian are almost exclusively preserved in these lithologies. Other aqueous minerals, including a wide variety of evaporite minerals (salts), and even ice, may also capture and preserve microorganisms. However, an important caveat for these classes of minerals is their residence time in the Earth's crust which tends to be comparatively short (100's of Ma for evaporites owing to dissolution and 100's of thousands of years for ice due to long-term climatic warming).

It is important to note that given the marked differences in environment and geological history, residence times for aqueous mineral deposits on Mars are likely to be quite different than on Earth. The surface hydrological cycle involving liquid water apparently died very early on Mars, and it is possible that evaporites formed in the waning stages of that early, wet period may still be preserved as surficial deposits on the floors of some paleolake basins (e.g. sites 147 and 148, Greeley and Thomas 1994). However, the same cannot be said for ice. The chaotic obliquity of Mars (Wisdom 1987) suggests that the present Martian chryosphere is likely to be very young owing to periodic global warming. Thus, Martian ice is unlikely to hold evidence of an early biosphere. There may be places on the surface, however, where ground ice may have been formed by the upflow of subsurface hydrological systems and thereby captured a subterranean microbiota, cryopreserving it in ground ice. The present Martian ice caps may also be an important archive of information about recent organic inputs from extrinsic sources, such as interplanetary dust particles or cometary impacts (Chyba and Sagan 1992; Chang 1995 ).

The basic criteria outlined above suggest that the long-term preservation of a fossil record on Mars is likely for only a restricted number of geologic environments. Aqueously deposited sediments in the oldest terranes on Mars, those formed during the early wet period, offer the greatest interest for exopaleontology. However, their discovery on Mars will require a more detailed knowledge of the surface geology and mineralogy of the Martian surface than we presently possess. We have yet to identify an aqueous mineral on surface of Mars with any degree of certainty. Thus, a first step in implementing a strategy for Mars exopaleontology is the identification of surficial aqueous mineral deposits that can be accessed by landed robotic missions.

As noted above, perhaps the most basic requirement to implement a strategy to explore for an ancient biosphere on Mars is the identification of key geologic environments and aqueously-deposited mineralogies from orbit (see Kerridge et al. 1995). In order for the proposed '05 sample return to legitimately address the concerns of exopaleontology, rock samples of appropriate mineralogy should be returned from a yet to be identified high priority site in the southern highlands of Mars. Given the coarse spatial resolution of the Thermal Emission Spectrometer (TES) which will be flown in 1996 (3 km/pixel; Christensen et al. 1992), it will likely prove difficult to resolve the precise spatial location of high priority mineral deposits. In addition, at the scale mapped, each pixel of TES data is likely to involve a complex mixture of mineralogies, and the deconvolution and isolation of discrete mineral spectra may likewise prove difficult or even impossible. Obviously, the solution to these problems is higher spatial resolution data from orbit.

Thus, to optimize site selection for exopaleontology in an '05 sample return, high resolution (100 m/pixel) compositional mapping at selected sites is deemed essential. And in order for high resolution mineralogical mapping to be available to assist in site selection for an '05 sample return, it must be obtained during the '01 opportunity. Clearly, high resolution data cannot be obtained for the entire planet, but with a well-conceived mission, it should be possible to build up high resolution mosaics of key geological sites at a scale that can isolate the spatial location of individual deposits and outcrops. This recommendation has been challenged by those who assume that the entire Martian surface is either blanketed by fine dust, or obscured by weathering rinds or "patinas". But this seems quite unlikely and is unsupported by the views of rocks obtained at the two Viking lander sites which show many rocks with fresh-looking surfaces.

The rocks at the surface of Mars are subjected to extreme changes in temperature on a diurnal and seasonal basis. On Earth such physical weathering processes involving temperature extremes in the absence of water, produces shattered and exfoliated outcrops. Fresh rock surfaces are constantly being exposed by such processes on Earth, and this is very likely to be the case on Mars as well. In addition to the ongoing physical weathering processes on Mars, the dominant erosional processes constantly break up rock materials during transport, exposing fresh surfaces. Wind erosion has exposed extensive areas of bedrock (as yardangs) and gravity, acting in combination with ice, has produced massive debris flows and rock glaciers in steeper terranes. Thus, at high spatial resolution, it is quite likely we will be able to obtain the strong primary signatures needed to map mineralogy and locate discrete deposits of interest.

High resolution data obtained from orbit will not only provide a basis for detailed site studies for future landed missions, but will also yield valuable information to assist in the interpretation of data obtained by the Thermal Emission Spectrometer during its global mapping exercise (e.g. deconvolution of mineral spectra and precise distributional data for deposits of interest). But it is unlikely that a mid-IR orbital instrument can achieve an acceptable signal to noise ratio from orbit within existing cost/weight constraints. Therefore it is recommended that high spatial resolution (100 m/pixel) data be obtained using a near-IR (1-5 um range), hyperspectral (10-20 nm bandwidths) imaging system to create maps of high priority target areas for future landed missions. The technologies needed to accomplish this task are relatively mature, as near-IR mapping has become a standard exploration tool in the minerals industry (Jonathan Huntington, pers. comm. 1996).

A proposal presented during the Mars Sample Return Workshop (Ronald Greeley, Arizona State University) involves a mid-latitude aerobot/balloon mission to Mars that would carry a high spatial resolution mid-IR spectral imager. Although this mission concept presently lies outside of the planned Mars Global Surveyor Program, it has recently gained prominence in mission planning activities since the March meeting (e.g. Roadmap for Solar System Exploration) and is discussed here more fully within the context of a potential exopaleontological mission.

The chief disadvantage of a balloon mission is that the remote sensing coverage obtained by a single balloon platform is restricted to a comparatively small band of latitude. However, because of the low altitude, such a mission could easily attain the desired 100 m/pixel spatial resolution using a mid-IR mapping spectrometer. Optimally, the spectral range of a mid-IR instrument carried on a balloon mission would be in the 5-12 um range where many fundamental vibrations of high priority aqueous minerals (e.g. carbonates, silica, evaporites) may be detected. Such a mission could deliver high spatial resolution data with the same spectral range as the global TES instrument. This would be extremely valuable in providing strips of very high resolution data at key sites to assist in interpreting the global data set obtained by the orbital TES mission in MGS '96.

It would be preferable from the standpoint of exopaleontology to deploy such a mission over southern highland terranes at a latitude that would transect as many high priority targets as possible during a nominal mission. The altitude should be optimized to provide the highest possible spatial resolution. In some scenarios, the balloon descends to the surface at night where a magnetic sond could be deployed to explore for near-surface ground ice. In that kind of scenario it would be optimal from an exopaleontological standpoint to also carry out mineralogical analysis of the surface at each descent to assist in the mapping and exploration for aqueous mineralogies. The disadvantage of such a mission would be the lack of control over where the balloon touched down on any given descent. However, model calculations indicate many traverses would be made around the planet during a nominal mission, so that a potentially much broader area and larger number of sites could be sampled within a single latitudinal band, than would otherwise be possible using long-ranging rovers. It should also be pointed out that balloon platforms could also be used to carry other types of remote sensing instruments of exobiological interest, such as high resolution imagers to detect point sources of methane, water vapor, or other potential exobiological "oases" on the surface.

To provide data that could be used to plan for an '05 sample return, a balloon mission would also need to fly during the '01 opportunity. It is unclear from a standpoint of technical readiness whether an '01 mission is feasible, but balloons provide one alternative solution to the problem of high spatial resolution mineralogical mapping, and could be valuable even if flown later in the exploration program.

Obviously, site recommendations for future landed opportunities will necessarily reflect a balance of the programmatic goals mentioned in the Introduction. The Global Surveyor Program, as presently defined, identifies the following exploration goals: the search for past and present Martian life, Mars volatile and climate history, and the distribution of resources (geological materials making up the Martian surface, including water). Clearly, the question of a past or present Martian biosphere is highly interdisciplinary in nature, and requires information about both the volatile and climate history, and surface geology. But there are several key recommendations for implementing the present strategy for the exobiological exploration of Mars (see Kerridge et al. 1995). These requirements include 1) discovering the precise location of aqueously-deposited minerals from orbit, (in particular, those that have a high priority for exopaleontology), 2) precisely delivering capable rovers to high priority geological targets on the Martian surface (those that have highest probability for having preserved biosignatures, and/or prebiotic organic compounds), and 3) carrying out detailed in situ mineralogical and organic assays at high priority locations in order to pre-select rock types before returning them to Earth (this is important because we cannot bring back every available rock sample at a site and must prioritize and target specific rocks prior to analysis or sample return).

As pointed out previously, the highest priority requirement for a Mars sample return in '05 or beyond is to find aqueously-deposited sediments of proper mineralogy and age. Future missions must get beyond simple elemental analysis and determine the precise structural configurations of molecular compounds required for mineralogical determinations. This is necessary because it is the structural configuration of elements making up minerals (and not the elements themselves) that carries information about past environments. Elemental abundances alone cannot provide the constraints needed to determine mineralogy because the same elements combine in different ways under environmental conditions. In addition, structural polymorphism (minerals with the same composition but different internal structures) are commonplace, each form being indicative of a different environment.

Exopaleontology's requirement for specific mineralogical information aligns very well with the objectives of other disciplines in Mars exploration, because the mineralogies that have been targeted for exopaleontology, are the same ones that hold the most information about climate and volatile history. In addition, aqueous minerals figure prominently in defining a resource base for such things as in situ oxygen, water and propellant production.

Because we cannot possibly expect to go just anywhere on the surface of Mars and find evidence of past or present life, exploration should proceed in step-wise fashion from orbital reconnaissance, to more targeted robotic surface exploration, progressively narrowing the search strategy leading up to sample return (see Kerridge et al. 1995). Beyond the currently-defined missions in '96, the next step in this process should be to obtain high spatial resolution mineralogical maps of key sites from orbit. This should be followed by surface missions that are capable of delivering long-ranging rovers to specific geological targets to obtain mineralogical and organic analyses. A basic requirement for all in situ analyses for mineralogy or organics is the need to acquire data from freshly exposed (interior) rock surfaces that are unweathered and dust-free. The third requirement is for the return of rock samples of appropriate aqueous mineralogy to Earth for detailed microscopic and geochemical analysis. The strategy for Mars exopaleontology (Farmer and Des Marais 1994; Farmer 1995) has identified several high priority geological targets that will require multiple landed missions and sample returns in order to fully address all potential past environments for life. This should be reflected in any exploration "roadmap" for Mars as multiple sample returns beyond the MGS '05 sample return.

To achieve a maximum science return for exopaleontology during a first sample return, the following milestones should be achieved during the precursor landed mission in '03:

Mobility: The rover in '03 should be capable of a multiple km traverse during the nominal mission in order to provide access to a broad sampling of geologic targets at a site of exopaleontological interest.

Sample Selection: Rovers in both '03 and '05 should be able to survey rock fields and pre-select individual target rocks for in situ analysis and (in '05) sample return, based upon their mineralogy. This will require high resolution visible range cameras and a rover-mounted (preferably mid-) IR spectrometer.

Microscopic Imaging: Once targets have been identified, rovers should be able to image weathered and fresh rock surfaces at 'hand lens' magnifications (0.1 mm resolution) in order to visualize microtextures of rocks. Optimally, illumination systems for rover hand-lenses would be able to deliver visible, infrared and UV wavelengths to assist in textural and compositional evaluations. UV could be particularly valuable because many minerals and organic materials flouresce with unique spectral signatures.

Access to Rock Interiors: Rovers should have ability to access rock interiors by exposing fresh surfaces, either through breakage or abrasion. This capability is regarded as a key requirement for all analytical tools that seek to evaluate composition.

In Situ Mineralogical Analysis: Although elemental analysis is regarded as important in providing information about the biogenic elements, in order to address the important issues of exopaleontology, instrument payloads must also provide information about molecular structure that will lead to an understanding of mineralogy. Key technologies for mineralogical analysis include 1) more qualitative methods of IR spectroscopy, Laser Raman, or reflected mode X-ray diffraction which require freshly exposed rock surfaces, and 2) the more definitive method of transmission mode X-ray Diffraction, which requires a powdered sample.

Redox and Organic Analysis: In order to optimize for the screening of rock samples most likely to contain organic matter, instrumentation should have the ability to determine the oxidation state of key elements, like iron. Mineralogies formed under oxidizing conditions are unlikely to preserve reduced organic compounds, although biofabric information may be present. The methods listed above provide primary mineralogical data that can be used to judge the redox environment for a broad variety of materials. In addition, Mossbauer spectroscopy is a more focused method for iron-bearing minerals. Once candidate mineralogies are discovered, new methods of laser-assisted spectroscopy can be used to directly detect organic compounds in rocks.

Because only selected areas will be mapped at high resolution during upcoming orbital missions, targets for high resolution imaging should include the highest priority sites for Exopaleontology. A catalog of such sites is presently being assembled to assist mission planners (e.g. Farmer et al. 1994). Although there are a number of geological site types of potential exopaleontological interest (Farmer and Des Marais 1994), the most easily identified targets from orbit are likely to be larger aqueously-deposited sediments on the floors of ancient paleolake basins in the southern highlands of Mars (e.g. Parana Vallis; Farmer et al. 1995). Such sites may provide access to a variety of aqueous mineralogies, including fine-grained detrital sediments, such as claystones, shales and marls, and chemical precipitates, including various evaporite deposits, spring-deposited carbonates, or more broadly-distributed sedimentary cements.

Landing sites for '03 and '05 should be chosen from a list of high priority exopaleontological sites that have been the focus of prior geological studies using high resolution orbital imaging and compositional data obtained during the '96 and '98 MGS missions. This indicates the importance of present site selection efforts using Viking data to define high priority exopaleontological sites for inclusion among the targets for high resolution imaging during the '96 mission.

As indicated previously, the preferred samples for exopaleontology are aqueously deposited sediments of low permeability that were formed rapidly either by precipitation or rapid detrital sedimentation. To preserve organic materials, host rocks need to be made up of stable minerals, such as silica or phosphate, capable of maintaining a closed chemical system during subsequent post-burial changes (diagenesis). Given the cold, dry conditions on the Martian surface, it is possible that even less stable lithologies (e.g. carbonates or evaporites which are prone to recrystallization or dissolution under more active hydrological systems and deeper burial conditions on Earth), will be well preserved on Mars. Thus, all fine-grained aqueous sediments should be considered prime targets for a sample return.

As noted, the optimal sample for Mars exopaleontology is an aqueously-deposited sedimentary rock of stable mineralogy from a site in the ancient highlands. Such a sample optimizes the opportunity to preserve biological or prebiotic organic signatures from that early period in Martian history when we believe liquid water existed at the surface of the planet, and when life may have been present. Because of the strong control that geologic environment exerts on the preservation of biosignatures in rocks, the optimal samples for exopaleontology are very sensitive to location. Therefore, site selection is of paramount importance. As a first step in implementing this strategy, a sample return mission in '05 should be carried out at a site which, based on orbital data from precursor missions, has a good chance of including aqueously-deposited sedimentary rocks of the right composition and texture.

Given the chaotic obliquity of the planet (Wisdom 1987), it is quite plausible that the climate of Mars has varied substantially over the millennia. Thus, a fine-grained soil sample from Mars is most likely to record a complex history of overlapping temporal and spatial events, mostly conditioned by the most recent climatic changes that have occurred on the planet. Although fine-grained soil samples may provide valuable information about recent climate and weathering processes, such samples are far less likely to have retained information about the environmental conditions that existed early in the history of Mars. Therefore such samples are not the most desirable for addressing the strategy formulated by the broader Exobiology community (see Kerridge et al. 1995).

Analysis of the materials making up soils at the Viking 1 and 2 landing sites suggested an average grain size of 4 microns. Most grains were inferred to be composite, and the actual grain size is probably much smaller than 4 microns. At such small grain sizes, the minerals present are mostly clays, or other weathering products, and in the case of Mars, mostly wind-transported clays from unknown source areas. The higher reactivities of clay-sized mineral grains favors the rapid attainment of chemical equilibrium under prevailing surface conditions. Such equilibration results in the loss of the primary mineralogical or geochemical information that can tell us about the original depositional environment.

To speculate that a Martian soil has retained information about early climate is to assume that 1) surface conditions have remained constant for billions of years, or 2) that the fine-grained materials comprising Martian soils have remained chemically inert for an equally long period of time. But even under the cold, dry conditions of the present Martian environment this seems untenable. The strong compositional similarity of Martian soils sampled at the two widely separated and geologically-distinct Viking landing sites, suggests that soils there are comprised of a globally-distributed, homogeneous aeolian material that has reached chemical equilibrium with present surface conditions.

As noted above, the fine components of Martian soils can be expected to retain little information about ancient depositional environments or processes. This size fraction of soils mostly consists of clays and other weathering products derived by the aqueous alteration of various parent rocks. In most soils, the types and compositions of clays or other minerals formed by weathering usually bear a complex relationship to their parent rocks, and interpretations about precursor materials are often poorly constrained, especially for wind-deposited fines which have been mixed and cannot be directly related to a source rock areas.

The ages of minerals making up fine-grained soil samples date the crystallization ages of various igneous terranes that were sediment sources, and not the time of deposition. Some authigenic minerals, such as glauconite, have been used to obtain depositional ages, but such minerals are notorius for being easily altered, and therefore subject to losses of parent of daughter product. This can alter the mineral age in an unpredictable way. Thus, Martian soils are likely to yield a broad spread of radiometric ages with 1) modes clustering near the average crystallization ages for various igneous minerals that may be present (reflecting the ages of original igneous source terranes sampled by the wind or other processes) and 2) modes reflecting the resetting of primary ages by shock or burial metamorphism, and losses of volatile parent isotopes or daughter products during younger periods of chemical weathering. Clearly, the age structure of such samples is complex, and poorly constrained. A composite age of such a sample is rarely useful for inferring historical events, and that is why such approaches are rarely used in geology. With care, ages obtained from separate igneous minerals in a sample (if they are still present) may yield broadly constrained maximum ages for potential source terranes, and the ages of some weathering products may cluster to indicate more recent periods of aqueous alteration and chemical weathering. Inferring age relationships from soils is not a straightforward task, however, and that is why such samples are rarely used in age dating.

Although the secondary mineral assemblages that make up present Martian soils may provide important information about recent climatic and weathering processes and the recent volatile history of Mars, such materials are quite unlikely to provide answers to the existence of past or present Martian life. The GCMS experiments of Viking indicated that the fine materials making up the highly oxidized soils of Mars do not retain any evidence of organic compounds, even though we should expect some background level of organics due to the constant rain of IDP's. The results of the Viking biology experiments, combined with the GCMS results indicate the presence of highly reactive soil oxidants that are destructive to organic molecules. As noted previously, the preservation of organic signatures in rocks requires that they be quickly sequestered by fine-grained, stable, aqueous mineralogies shortly after they are introduced into the environment, to protect them from oxidation. The selection of a site where such processes could have occurred should be considered a first step in sample return scenarios that want to address exopaleontological issues. Coarse sand to pebble-sized grains of aqueous lithologies in soils may retain biomolecular and biogeochemical information, but the probability of retention decreases with decreasing grain size because of the tendency of small grains to equilibrate to existing conditions in the soil. However, supporting microfabric information, crucial for inferring depositional origin, is typically not retained in even sand-sized grains. On that basis, pebble-sized grains of aqueous lithologies (e.g. dense rocks like cherts, phosphates, and carbonates) in soils should be considered a minimal requirement for a sample return that wishes to address exopaleontological issues. Pebble sized grains afford enough protection to preserve organics from oxidation, while providing the proper scale for preserving and visualizing biosedimentary fabrics. Thus, soils with pebble-sized clasts of appropriate mineralogy provide a compromise position for addressing the preservational concerns of exopaleontology.

In summary, a grab sample of fine-grained soil in '05 would probably advance our understanding of modern climate and weathering processes on Mars, and provided it contained mineral grains of primary igneous lithologies, could yield broadly constrained information about the maximum age of the Martian surface. However, age relationships obtained from soils are likely to be complex and difficult to interpret. Grab samples of fine-grained soils are unlikely to provide any answers to the question of Martian life, and in that sense such sample return scenarios do not represent an optimal strategy for exopaleontology. At this stage in our knowledge, the optimal sample return for exopaleontology should include samples of aqueously-deposited sedimentary rocks. The inclusion of even small pebbles of aqueous sedimentary rocks in a returned soil sample would greatly enhance its value for exopaleontology, and for studies of ancient climate and volatiles as well.

To optimize the proposed '05 Mars sample return for exopaleontology, missions in the '01-'05 time frame should be structured to address the following key issues:

1) High resolution, near-IR spectroscopy should be during the '01 orbital mission to ensure the selection of the best site for a sample return in '05.

2) Landed missions in '03 and '05 should be targeted to sites with evidence of aqueously-deposited sedimentary rocks exposed at the surface. A favored site for the first sample return mission is the floor of a paleolake basin in ancient highland terrane where fine-grained chemical or detrital sediments could be present at the surface. Many paleolake basins are large features that provide more accessible targets both for orbital reconnaissance and landed missions. However, the site selection criteria should be re-evaluated as compositional data becomes available from the '96.

3) The landed mission in '03 should include a long-ranging rover capable of traversing multiple kms during a nominal mission. This will help to ensure sampling of a broad variety of lithological types at the landing site.

4) The rover in '03 should be able to survey rock fields to qualitatively evaluate composition, and then pre-select rocks for more detailed, rover-based analysis using instruments capable of identifying mineralogy and organic constituents. Instrumentation for detailed analysis should focus on methods involving reflected energy (instruments placed against rock surface) to avoid complex sample preparation and delivery systems.

5) Methods used for mineralogical or organic analysis, all require access to fresh (interior) rock surfaces free of dust coatings or patinas (weathering rinds). Thus, a very high priority is placed on the ability of a rover to prepare a rock surface by breakage or abrasion.

6) The optimal samples for exopaleontology are rocks deposited in aqueous environments. Grab samples of fine soils or related materials formed by secondary weathering processes may provide important information about the recent climate and volatile history. However, it is important to realize that such samples are unlikely to provide information about the existence of past or present Martian life, and should not be viewed as an exobiological mission. The primary value of a fine-grained Martian soil sample will be in improving our understanding recent hydrological and weathering processes, and possibly some absolute age dating of terranes. In that sense, such a sample holds value for the fundamental issue of the history of Martian water, fundamental for evaluating the potential for past of present life. The inclusion of small rocks of aqueous mineralogy in a Martian soil sample would greatly enhance its value for exopaleontology. A critical issue in that regard is the selection of a sample return site likely to have aqueous lithologies exposed at the surface.

7) Balloon missions offer an alternative way to obtain high resolution remote sensing data of interest to exopaleontology. Balloon missions at mid-southern latitudes would hold the greatest interest for exopaleontology. Although the remote sensing data obtained from any given balloon mission will be restricted to a band of latitude, this disadvantage can be offset by the potential for descents at many more locations on the surface than would be possible by rovers. However, balloon mission scenarios that incorporate regular descents to the surface should carry instrumentation for in situ mineralogical analysis and the detection of ground ice. To assist in the search for near-surface oases for extant life, balloon missions should carry high spatial resolution remote sensing instruments capable of detecting water or methane.

8) Given that there are a number of potential geological environments and lithological targets for a fossil record on Mars, it is unlikely that the question of Martian life can be answered with a single sample return mission in '05. We therefore recommend a continuing program of landed mission and sample returns after '05 to systematically address the comparatively small number of site types that hold the greatest potential for having preserved a record of ancient life, or a prebiotic organic record.

REFERENCES

Boston, P.J., M.V. Ivanov, C.P. McKay. 1992. On the possibility of chemosynthetic ecosystems in subsurface habitats on Mars. Icarus 95: 300-308.

Carr, M.H. 1996. Water on Mars. Oxford Univ. Press 229 p.

Chang, S. Prebiotic chemistry on Mars. Eos, Transactions, American Geophysical Union, 76 (46): F334. 1995

Christensen, P. R., D. L. Anderson, S. C. Chase, R. N. Clark, H. H. Kieffer, M. C. Malin, J. C. Pearl, J. Carpenter, N. Bandiera, F. G. Brown and S. Silverman 1992. Thermal emission spectrometer experiment: Mars Observer Mission. Journal of Geophysical Research 97, 7719-7734.

Chyba, C.F. and C. Sagan. 1992. Endogenous production, , exogenous delivery, and impact-shock synthesis of organic molecules: An inventory for the origins of life. Nature 355: 125-131.

Farmer, Jack D. 1995. Mars Exopaleontology. Palaios 10 (3): 197-198.

Farmer, J., Des Marais, D., Greeley, R., Landheim, R. and Klein, H. 1995. Site selection for Mars Exobiology. Advances in Space Research 15(3): (3)157-(3)162.

Farmer, J., and D. Des Marais. 1994. Exopaleontology and the search for a fossil record on Mars. Lunar Planetary Science 25: 367-368.

Farmer, J., Des Marais, D., Klein. 1994. Mars Site selection for Exobiology: Criteria and Methodology. pp. 11-16. In Greeley, R. and P. Thomas (ed.). Mars Landing Site Catalog. (Second Edition) NASA Reference Publication 1238, 392 p.

Greeley, R. and P. Thomas (ed.). 1994. Mars Landing Site Catalog. (Second Edition) NASA Reference Publication 1238, 392 p.

Kerridge, J. (ed.) 1995. An Exobiological Strategy for Mars Exploration. NASA Special Publication 530, 55 pp.

McKay, C.P. and C.R. Stoker. 1989. The early environment and its early evolution on Mars: Implications for life. Rev. Geophys. 27, 189-214.

Wisdom, J. 1987. Urey Prize lecture. Chaotic dynamics in the solar system. Icarus 72: 241-275.

NASA     Space Science    Solar System Exploration     Exobiology Program     Exobiology Branch at ARC