To be published in: Astrobiology's Last Frontiers: Distribution and Destiny of Life in the Universe, in "Origins: Genesis, Evolution and the Biodiversity of Life ", ed. J. Seckbach in the COLE series, Vol. 6, Kluwer Academic Publishers, 2003, Dordrecht, by invitation.

ASTROBIOLOGY'S LAST FRONTIERS:
Distribution and destiny of life in the universe.

JULIAN CHELA-FLORES
The Abdus Salam
International Centre for Theoretical Physics
P.O.Box 586; Strada Costiera 11;
34136 Trieste, Italy and
Instituto de Estudios Avanzados,
Apartado Postal 17606 Parque Central,
Caracas 1015A, Venezuela.

1. Introduction

The four areas that define the new science of astrobiology are the origin, evolution, distribution and destiny of life in the universe. It is undoubtedly the fourth one which is most likely to encourage interdisciplinary dialogue. This is not only in relation with the more evident physically- and biologically-oriented disciplines, but also with other not so close, but yet significant branches of culture. Indeed, the destiny of life in the universe will depend on the cosmological model that will eventually be best supported by experiments in the high-energy physics domain; it will also depend on observation of the largest red-shifts of receding galaxies. The case of an ever-expanding universe is an evident example of great interest to several areas of culture; for if there would not be a reversal of the big bang expansion (which seems to have some data in its favor), the possibility of the phenomenon of life surviving on a large enough scale of time seems unlikely.
On the other hand the distribution of life in the universe, in spite of lacking solid theoretical, or observational bases, has in its favor that it is an aspect of astrobiology that can be probed in terms of a wide range of projects and space missions especially dedicated to explore the question of whether or not we are alone in the universe. In the companion paper (Chela-Flores, 2002) we have persevered with the hypothesis that was first formulated during the commemoration of the centenary of Charles Darwin's death (Dawkins, 1982). The points made on that occasion are relevant in the context of this paper, since Darwin's theory of evolution is assumed to be the only theory that can adequately account for the phenomena that we associate with life anywhere in the universe. In the present paper we examine the question whether evolution of intelligent behavior is just a matter of time and preservation of steady planetary conditions, and hence ubiquitous in the universe. This question is motivated by the problem of understanding the bases on which we can get significant insights into the question of the distribution of life in the universe. Such information would also have deep implications on the other frontier of astrobiology, the destiny of life in the universe. In Secs. 2-4 arguments are presented within the theory of evolution that lead us to a rationale, within the life sciences, for testing the universality of Darwin's theory.

2. To what extent is natural selection stronger than contingency?

2.1 NATURAL SELECTION VERSUS CONTINGENCY

For simplicity, in Sec. 4 we shall focus on various degrees of convergence across all scientific disciplines that are relevant to astrobiology. Features which become more, rather than less similar through independent evolution, will be called 'convergent'. In fact, convergence in biology is often associated with similarity of function, as in the evolution of wings in birds and bats. An example is provided by New World cacti and the African spurge family, such as some euphorbs (for example, the Euphorbia stapfii), and even some members of the Madagascar Didieraceae (Didiera madagascariensis). These plants are similar in appearance, being succulent, spiny, water-storing, and adapted to desert conditions (Tudge, 1991; Nigel-Hepper, 1982). However, they are classified in separate and distinct families, sharing characteristics that have evolved independently in response to similar environmental challenges, and hence this is a typical case of convergence.
On the other hand, adaptive radiation is a second Darwinian concept that will be necessary for understanding the examples in this section that will argue in favor of restricted predictability in astrobiology. In fact, adaptive radiation simply means evolution of an animal, or plant group into a wide variety of types adapted to specialized modes of life. In other words, adaptive radiation signifies evolutionary diversification of a single lineage into a variety of species with different adaptive properties. Darwin's finches provide the classical example of adaptive radiation. Thirteen species of Darwin's finches live in the Galapagos Islands. They differ in the shape of their beaks.
It is remarkable how versatile their beaks can be: keratin, the substance form which they are made, can be easily molded by evolutionary pressures, thereby facilitating the origin of all the species now inhabiting these islands. (Besides, there is an additional species inhabiting Cocos Island in the Costa Rican territorial waters in the Pacific ocean, north of the Galapagos Islands.) We shall consider below, some less familiar examples of adaptive radiation that will argue in favor of certain degree of predictability in biology (and clearly in all its branches, particularly in astrobiolgy, in view of the assumed universality of Darwinism).
Before we approach the central question of convergent evolution, however, we should recall, firstly that so far factors influencing the relative degree to which the Earth biota has been shaped are still a debatable topic. According to the hypothesis of universal Darwinism (cf., Sec. 1), life on Earth, and elsewhere, may have been shaped mainly by:
(i) contingency, or by
(ii) the gradual action of evolution's main driving force (i.e., natural selection).
However, it may be possible to ascertain experimentally whether, independent of historical contingency, natural selection is powerful enough for organisms living in similar environments to be shaped to similar ends. For this reason, it is important to highlight the following three recent examples, which suggest that, to a certain extent and in certain conditions, natural selection may be stronger than chance:
o Sticklebacks (Gasterosteus spp.) are members of the fish Gasterosteide Family; they are, in fact, small northern hemisphere fishes with spines in front of their dorsal and ventral fins. Empirical evidence has been presented (Rundle et al., 2000), which demonstrates that natural selection plays a fundamental role in the early stages of the evolution of new species. The case in question concerns a striking example of convergent evolution: sticklebacks were originally of marine origin (G. aculeatus), but were trapped in three different lakes on the Canadian Pacific coast (Lakes Priest, Enos and Paxton) at the end of the last ice-age, as the glaciers retreated. In spite of the fact that the lakes have remained isolated, the same two species have formed in all three lakes.
Two non-interbreeding varieties have evolved in each lake; firstly, a bulky bottom-dwelling (benthic) type and, secondly, streamlined actively swimming individuals (limnetic), which feed in the open water. The mating preferences of the fish were tested. It was found that benthic mated with benthic, both from their own lake and form the other two. In addition, it was found that limnetic mated with limnetic. This remark demonstrates that natural selection has been the driving force in the evolution that has taken place in the above three Canadian lakes.
o Black European fruit flies (Drosophila subobscura) were transported to California over twenty years ago. This event has provided the possibility of testing the role of natural selection in two different continental environments. Pacific coast D. subobscura (Santa Barbara to Vancouver) were compared in wing-length with European ones (from Southern Spain to the middle of Denmark). After half a dozen generations were observed in similar conditions, the increase in wing length was almost identical (4%). This is a compelling case in favor of the key role played by natural selection in evolution (Huey et al., 2000).
o Anole lizards from some Caribbean islands (Anolis spp.) provide another example of evolutionary convergence. The islands are Cuba, Hispaniola (shared by Haiti and the Dominican Republic), Jamaica and Puerto Rico (the so called Greater Antilles). The observed phenomenon suggests that in similar environments adaptive radiation can overcome historical contingencies in order to produce strikingly similar evolutionary outcomes. We could even say that there has been replicated adaptive radiation in the various islands.
In fact, what has been shown (Losos et al., 1998) is that although it had been known for some time that dozens of species thrive on these islands, some groups of lizards from different islands living in similar environments also look similar. Genetic analysis has shown that similar traits have evolved in distantly related species for coping with similar environments (such as tree-tops or ground-dwelling): anoles that live on the ground have long, strong hindlegs, while those living at tree tops have large toe-pads and short legs. Repeated evolution of similar groups of species (both morphologically and ecologically) suggests that adaptation is responsible for the predictable evolutionary responses of the anole lizards of the Caribbean. Indeed, we can speak in this case of evolutionary history repeating itself (Vogel, 1998).
These arguments support the following views:
(a) It is necessary to argue that within certain limits the outcome of evolutionary processes might be rather predictable (Conway Morris, 1998).
(b) Certain directions of evolution may carry such decisive selective advantages as to have high probability of occurring elsewhere [in the universe] as well (De Duve, 1995).
(c) The ubiquity of evolutionary convergence argues against the view that biological diversity on this planet is unique to Earth (Conway Morris, 2002).

2.2 CONSTRAINTS ON CHANCE

It is instructive to appreciate that several constraints on chance are relevant to the question of whether life elsewhere might follow pathways analogous to the transition terrestrial ones. Various examples of constraints on chance have been enumerated elsewhere (De Duve, 1995). We shall comment on them and later we shall provide an additional example:
o Not all genes are equally significant targets for evolution. The genes involved in significant evolutionary steps are few in number, they are the so called regulatory genes. In these cases mutations may deleterious and are, therefore, not fixed.
o Once a given evolutionary change has been retained by natural selection, future changes are severely constrained; for example, once a multicellular body plan has been introduced, future changes are not totally random, as the viability of the organisms narrows down the possibilities. For instance, once the body plan of mammals has been adopted, mutations such as those that are observed in Drosophila, which exchange major parts of their body, are excluded. Such fruit-fly mutations are impossible in the more advanced mammalian body plan.
o Not every genetic change retained by natural selection is equally decisive. They may lead more to increasing biodiversity, rather than contributing to a significant change in the course of evolution. This may be illustrated within the Solanaceae family (Brown, 1966): one tomato chromosome has a region between its center and end ('centromere' and 'telomere'), which consists of a row of segments in which DNA is compacted into tight masses, largely inactive in transcription ('chromomeres') in Petunia, in spite of being another genus of the same family, the abundance of chromomeres is not preserved, since larger blocks of heterochromatin are observed. A tiny bit of heterochromatin may be superficially indistinguishable from a eukaryotic 'chromomere'. These two genera illustrate how quickly evolution can induce rearrangements of heterochromatin, while preserving general chromosome structure; this mutation has not contributed any significant change in the course of evolution.
Besides the constraints on chance mentioned in the last section, we should recall the eternal confrontation deep in the fabric of evolutionary theory, brought to popular attention by Jacques Monod in his book Chance and Necessity (Monod, 1972).
Indeed, implicit in Darwin's work we have chance represented by the randomness of mutations in the genetic patrimony, and their necessary filtering by natural selection. However, the novel point of view that astrobiology forces upon us accepts that randomness is built into the fabric of the living process. Yet, contingency, represented by the large number of possibilities for evolutionary pathways, is limited by a series of constraints, as mentioned before. What we feel is even more significant for astrobiology, is to recognize that natural selection necessarily seeks solutions for the adaptation of evolving organisms to a relatively limited number of possible environments. We will see in cosmochemistry (cf., Sec. 4.1) that the elements used by the macromolecules of life are ubiquitous in the cosmos.
To sum up, the finite number of environments forces upon natural selection a limited number of options for the evolution of organisms. From these remarks we expect convergent evolution to occur repeatedly, wherever life arises. It will make sense, therefore, to search for the analogues of the attributes that we have learnt to recognize in our own particular planet.

3. If local environments can shape how organisms change with time through natural selection, how frequent are cosmic environments favorable to life's origin?

In the original theory of Darwin the possibility had been raised that local environments shape how organisms change with time through natural selection. In view of the evidence presented in Secs. 1 and 2, we assume that natural selection is the main driving force of evolution in the universe. For these reasons it is relevant to question whether local environments that were favorable for the emergence of life on the early Earth, were at all unique, occurring exclusively in our own solar system.
Alternatively, we may question, as we do in the present paper, whether other environments fulfill conditions favorable to life's origin, either within our solar system, or in any of the planets, or satellites, in the multiple examples of solar systems known at present. In addition, we suppose that on such bodies steady conditions are preserved. By steady conditions it should be understood that the planet where life may evolve is bound to a star that lasts long enough: in other words, the time available for the origin and evolution of life should be sufficient to allow life itself to evolve before the solar system of the host planet, or satellite, reaches the final stages of stellar evolution, such as the red-giant and supernova phases. It is also assumed that major collisions of large meteorites with the world supporting life are infrequent after the solar system has passed through its early period of formation.
In such steady conditions, the gradual action of natural selection will be expected to be the dominant mechanism in evolution, in view of the assumed universality of Darwinism (cf., Sec. 1). Besides, convergence is expected to occur at various levels: in prebiotic evolution (cf., Sec. 4.1), chemical evolution (cf., Sec. 4.7), and finally, in biological evolution (cf., Sec. 4.4 and 4.5). Fortunately, the existence of steady Earth-like planetary conditions is an empirical question for which we will be able to give partial answers in the foreseeable future; observational means will be provided by progress in space interferometry. One possibility for achieving this objective will be the Darwin Project (Fridlund, 2001).

4. Evolutionary convergence beyond the framework of biology

The above examples suggest that natural selection is powerful enough to shape organisms to similar ends, independent of contingency. We shall enumerate examples of general aspects of convergence at different levels of evolution: cosmic, planetary, molecular, biochemical and, once again, biological.

4.1 COSMIC CONVERGENCE

Hydrogen and helium make up almost the totality of the chemical species of the Universe. Only 2% of matter is of a different nature, of which approximately one half is made by the five additional biogenic elements (C, N, O, S, P). From organic chemistry we know that nuclear synthesis is relevant for the generation of the elements of the Periodic Table beyond hydrogen and helium and, eventually, for the first appearance of life in solar systems. The elements synthesized in stellar interiors are needed for making the organic compounds that have been observed in the circumstellar, as well as in interstellar medium, in comets, and other small bodies. The same biogenic elements are also needed for the synthesis of biomolecules of life (cf., Sec. 4.7, "Convergence in Biochemistry").
Besides, the spontaneous generation of amino acids in the interstellar medium is suggested by general arguments based on biochemistry: the detection of amino acids in the room-temperature residue of an interstellar ice analogue that was ultraviolet-irradiated in a high vacuum has yielded 16 amino acids, some of which are also found in meteorites (Muñoz Caro et al., 2002; cf., also Bernstein et al., 2002).

4.2 CONVERGENCE IN THE FORMATION OF SOLAR SYSTEMS

Our solar system formed in the midst of a dense interstellar cloud of dust and gas. This event may have been triggered by the shock-wave of a supernova explosion. Indeed, there is some evidence for the presence of silicon carbide (carborundum, SiC) grains in the Murchison meteorite, where isotopic ratios demonstrate that they are matter from a type II supernova (Hoppe et al., 1997). In the case of our solar system this occurred 4.6 billion years before the present (Gyr BP). We may be observing such a circumstellar disk around a young sun-like star (3 million years, Myr, of age) in the constellation of Monoceros:
A spinning cloud around the young star is having the brightness of its light regularly faded (it goes dim for 18 days every 48.3 days) by the interference of stellar photons and the cloud itself (Kerr, 2002). Several earlier examples of circumstellar disks are known, including a significantly narrow one around an 8 Myr old star. (The narrowness suggests the presence of planets constraining the disk.) The observation was by means of a spectrometer on board of the Hubble Space Telescope (Schneider et al., 1999).
The matter of the original collapsing interstellar cloud that does not coalesce into the star, collapses into a spinning circumstellar disk, where planets are thought to be formed in a process of accretion (some planetesimals collide and stay together, due to the gravitational force). In addition, a variety of small bodies are formed in the disk, prominent amongst which are comets, asteroids and meteorites.

4.3 CONVERGENT ORIGINS OF HYDROSPHERES AND ATMOSPHERES

Collisions of comets are thought to have played a role in the formation of the hydrosphere and atmosphere of habitable planets, such as the Earth. (This question may be settled in the next few years by the fleet of space missions that will interact with a variety of comets.) An alternative scenario supposes that volatiles emerged form the planet's interior through volcanic vents. The source of comets is the Oort cloud and Kuiper belt. These two components of the outer solar system seem to be common for other solar systems.
Hence we can also recognize evolutionary convergence in this cosmic sense. But there are additional factors, which contribute to the formation of habitable planets. We have already mentioned meteorites in the context of the formation of solar systems. In fact, the Murchison meteorite may even play a role in the origin of life: According to chemical analyses, some amino acids have been found in several meteorites: in Murchison we find basic molecules for the origin of life such as lipids, nucleotides, and over 70 amino acids (Cronin and Chang, 1993). Most of the amino acids are not relevant to life on Earth and may be unique to meteorites. This remark demonstrates that those amino acids present in the meteorite, which also play the role of protein monomers, are indeed of extraterrestrial origin.). In addition, chemical analysis has exposed the presence of a variety of amino acids in the Ivuna and Orgueil meteorites (Ehrenfreund et al., 2001). If the presence of biomolecules on the early Earth is due in part to the bombardment of interplanetary dust particles, or comets and meteorites, then the same phenomenon could have taken place in any of other solar systems.

4.4 CONVERGENCE AT THE LEVEL OF INVERTEBRATES

The evolutionary biology of the Bivalvia, both at the level of zoology and paleontology, provide multiple examples of convergence and parallel evolution, a fact that makes difficult the interpretation of their evolutionary history (Harper et al., 2000). Specific examples of convergence in mollusks have been pointed out earlier in the case of the camaenid, helminthoglyptid and helcid snails (Chela-Flores, 2001).

4.5 CONVERGENCE AT THE LEVEL OF VERTEBRATES

The examples of sticklebacks and anole lizards (cf., Sec. 2) provide two additional examples of evolutionary convergence in the vertebrates. Earlier we pointed out further examples amongst the vertebrates: Passeriformes is a group of birds which is often confused with Apodiformes, but are not related to them. Since this example of swallows and swifts constitutes a classical example of evolutionary convergence, we need not repeat it here (Chela-Flores, 2001).

4.6 CONVERGENCE AT THE MOLECULAR BIOLOGY LEVEL

Convergent evolution is manifest at the active sites of enzymes, in whole proteins, as well as in the genome itself, as we proceed to show:
o The lipases and even by the serine proteases. They have identical active sites (histidine, serine, aspartate form a catalytic triad). On the other hand, their folding is completely different (Tramontano, 2002).
o The northern sea cod (Boreogadus saida, Svalbard Norway) is an economically important marine fish of the family Gadidae. It is found on both sides of the North Atlantic. The distantly related order Perciformes with its suborder Percoidei, contains the sea basses, sunfishes, perches, and, more relevant to our interest, the notothenioid fishes from the Antarctic (Dissotichus mawsoni, McMurdo Sound, Antarctica). In spite of their distant relationship with cods, they have evolved the same type of antifreeze proteins, in which there are repeats of the amino acids threonine, alanine and proline (Chen et al., 1997.) These proteins are active in the fish's blood and avoid freezing by preventing the ice crystals from growing larger. The Antarctic fish protein arose over 7 million years BP, while the Arctic cod first appeared about 3 million years BP (both species arose in different episodes of genetic shuffling).
o The blind cave fish Astyanax fasciatus are sensitive to two long wavelength visual pigments. In humans the long wavelength green and red visual pigments diverged about 30 Myr BP. The mammalian lineage diverges form fishes about 400 Myr BP, but a recent episode in evolution has granted fish multiple wavelength-sensitive green and red pigments. Genetic analysis demonstrates that the red pigment in humans and fish evolved independently from the green pigment by a few identical amino acid substitutions (Yokoyama and Yokoyama, 1990), a clear case of evolutionary convergence at the molecular level.

4.7 CONVERGENCE IN BIOCHEMISTRY

The universal nature of biochemistry has been discussed form the point of view of the basic building blocks (Pace, 2001). One of the main points made in that paper is that it seems likely that the basic building blocks of life anywhere will be similar to our own. Amino acids are formed readily from simple organic compounds and occur in extraterrestrial bodies such as the above example of meteorites (cf., the Murchison meteorite in the section on "Convergent origins of hydrospheres and atmospheres"). Themes that are suggested to be common to life elsewhere in the cosmos are the capture of adequate energy from physical and chemical processes to conduct the chemical transformations that are necessary for life: lithotropy, photosynthesis and chemosynthesis. Other factors that militate in favor of the universality of biochemistry are physical constraints (temperature, pressure and volume), as well as genetic constraints.

5. Discussion: Can history repeat itself?

Ever since the publication of The origin of species, it has been argued that the possible course of evolution may be dominated by either of the above-mentioned alternatives [cf., Sec. 2, (i) and (ii)]. Random gene changes accumulating over time may imply that the course of evolution is generally unpredictable over time. But some care is needed in this assertion: What is certainly unpredictable is the future of a given lineage. This is due to the strong role in shaping life's evolutionary pathways played by contingent factors, such as extinction of species due to asteroid collisions with a given inhabited world, or other calamities. However, such uncertainties are of lesser interest to the larger issues that are relevant to astrobiology, namely the inevitability of the appearance of biological features, such as vision, locomotion, nervous systems, brains and, consequently intelligent behavior.
We have argued that contingency does not contradict a certain degree of predictability of the eventual biological properties that are likely to evolve. We should underline "biological property", as opposed to a "lineage", which is clearly a strongly dependent on contingency. In the companion paper we have sketched preliminary suggestions that would test whether the early stages of evolution of intelligent behavior are being driven mainly by natural selection at the cellular level in the context of planetary science (Chela-Flores, 2002).
If biology experiments succeed, such as the ones that are being planned in the near future, for instance in the context of the Mars Express mission and its lander Beagle-2, or even further ahead in time with missions to the Jovian moons, such successes would undoubtedly have a significant impact in our culture, not just in our scientific outlook. The influence of the new knowledge gained in astrobiology will also be felt while discussing deep philosophical questions, but also the influence will be felt in natural theology. These questions may be discussed not just against a background of our particular evolutionary line, which has been followed up by life on Earth. Such questions ought to be discussed already in terms of the many evolutionary lines that are hinted at by present-day developments in astrobiology. In fact, the intercultural dialogue is urgent for various reasons, which are strongly suggested by astrobiology:
o A human-level type of intelligent behavior may be widespread in the cosmos,
o Within the context of astrobiology it is clear that our human descent does go back all the way to microorganisms and,
o ultimately, our origins go back to star dust.
Yet, these three items lie outside our present culture, in spite of much current work in that direction. But it is not an easy task to integrate our knowledge, particularly with so much specialization in the nature of current research. Besides the general public finds it difficult to interpret so much information which is already available in the present age of informatics. In this work, my main overall thesis has been that such an integration is not only possible, but it is also timely, and indeed necessary.
Strictly speaking, a certain degree of predictability in astrobiology is not on the same footing as predictability in either physics (for instance, the exact time for an eclipse), or chemistry (for example, the result of a chemical reaction).
It is relevant to keep in mind that from neuroscience other aspects of the limited predictability of biology have been pointed out for some time (Manger et al, 1998). In fact, the tremendous expansion of the cetacean neocortex, while preserving the basic modular subdivisions (analogous to small-brained mammals such as the mouse) has led to a conjecture:
o There is a limited set of underlying mechanisms that are accessed to building brains.
Thus in the course of evolutionary history repetition of similar structures occurs. This also suggests that module size is evolutionary stable across species. Indeed with Krubitzer we can even question whether species differences are really so different (Krubitzer, 1995). In other words, in order to understand the evolution of the cortex and how organisms increase in behavioral complexity, it may be necessary to assume that anatomical alterations are generated from similar mechanisms.
Further, while the future of a given lineage cannot be predicted with certainty, the conjectured common mechanisms of cortex evolution, give us some certainty of the types of modification that are likely to occur in brain evolution. This aspect of brain evolution gives us some confidence that to a certain extent evolutionary history repeats itself in the context of the neurosciences, as in the earlier examples form the evolution of multicellular organisms (cf., Sec. 2). However, from what we have explained above, such limited predictability is nevertheless the aspect of evolution that is most relevant to astrobiology in relation to the question of life in the solar system, as well as in the wider context of evolution of intelligent behavior in the universe.

6. References

Bernstein, M.P., Dworkin, J.P., Sandford, S.A., Cooper, G.W. and Allamandola, L.J. (2002) Racemic amino acids from the ultraviolet photolysis of interstellar ice analogues. Nature 416, 401-403.

Brown, S.W. (1966) Heterochromatin. Science 151, 417-425.

Chela-Flores, J. (2001) The New Science of Astrobiology From Genesis of the Living Cell to Evolution of Intelligent Behavior in the Universe. Kluwer Academic Publishers: Dordrecht, The Netherlands, p. 161.

Chela-Flores, J. (2002) Can evolutionary convergence be tested on Europa? ESA SP 518. (To be published.)

Chen, L., DeVries, A.L. and Cheng, C-H. C. (1997) Convergent evolution of antifreeze glycoproteins in Antarctic notothenioid fish and Arctic cod. Proc. Natl. Acad. Sci. USA 94, 3817-3822.

Conway Morris, S. (1998) The Crucible of Creation. The Burgess Shale and the Rise of Animals, London, Oxford University Press, p. 202.

Conway Morris, S. (2002) First Steps towards defining galactic niches. Paper presented at the IAU Symposium 213 Bioastronomy 2002 Life Among the Stars. Summary of Proceedings, Hamilton Island, Great Barrier Reef, Australia July 8-12, Australian Centre for Astrobiology, Sydney, p. 12.

Cronin, J. R. and Chang, S. (1993) Organic matter in meteorites: Molecular and isotopic analyses of the Murchison meteorite, in The chemistry of life's origins. Greenberg, J. M., Mendoza-Gomez, C. X. and Pirronello, V. (eds.), Dordrecht: Kluwer Academic Publishers, pp. 209-258.

Dawkins, R. (1983) Universal Darwinism, in Evolution from molecules to men, Bendall, D.S. (ed.), London, Cambridge University Press, pp. 403-425.

De Duve, C. (1995) Vital Dust. Life as a cosmic imperative, New York, Basic Books, A Division of HarperCollins Publishers, pp. 296-297.

Ehrenfreund, P., Glavin, D.P., Botta, O., Cooper, G. and Bada, J.L. (2001) Extraterrestrial amino acids in Orgueil and Ivuna: Tracing the parent body of CI type carbonaceous chondrites. Proc. Natl. Acad. Sci. USA, 98, 2138-2141.

Fridlund, M. (2001) The Darwin Mission and Exo-/Astrobiology. ESA SP 496, 267-271.

Harper, E.M. Taylor, J.D. and Crame, J.A. (2000) The evolutionary biology of the Bivalvia, Geological Society Special Publication 177, London. www.palaeo-electronica.org/2001_2/books/evo.pdf

Hoppe, P., Strebel, R., Eberhadt, P., Amari, S. and Lewis, R.S. (1997) Type II supernova matter in a silicon carbide grain from the Murchison meteorite. Science 272, 1314-1317.

Huey, R., Gilchrist, G., Carlson, M., Berrigan, D. and Serra, L. (2000) Rapid evolution of a geographic cline in size in an introduced fly. Science 287, 308-309.

Kerr, R. A. (2002) Winking star unveils planetary birthplace. Science 296, 2312-2313.

Krubitzer, L. (1995) The organization of neocortex in mammals: are species differences so different? Trends in Neurosci., 18, 408-417.

Losos, J. B., Jackman, T.R., Larson, A., de Queiroz, K. and Rodriguez-Schettino, L. (1998) Contingency and determinism in replicated adaptive radiations of island lizards. Science 279, 2115-2118.

Manger, P., Sum, M., Szymanski, M., Ridgway, S. and Krubitzer, L. (1998) Modular subdivisions of dolphin insular cortex: Does evolutionary history repeats itself? J. Cognitive Neurosci., 10, 153-166.

Monod, J. (1972) Chance and Necessity An Essay on the Natural Philosophy of Modern Biology, Collins, London, p. 136.

Muñoz Caro, G.M. , Meierhenrich, U.J. , Schutte, W.A., Barbier, B., Arcones Segovia, A., Rosenbauer, H., Thiemann, W.H.P., Brack, A., Greenberg, J.M. (2002) Amino acids from ultraviolet irradiation of interstellar ice analogues. Nature 416, 403-406.

Nigel-Hepper, F. (1982) Kew: Gardens for Science and Pleasure, London, Her Majesty's Stationary Office, p. 81.

Pace, N. R. (2001) The universal nature of biochemistry. Proc. Natl. Acad. Sci. USA 98, 805-808.

Rundle H. D., Nagel, L., Wernick Boughman, J. and Schluter, D. 2000, Natural selection and parallel speciation insympatric sticklebacks. Science 287, 306-308.

Schneider, G., Smith, B. A., Becklin, E. E., Koerner, D. W., Meier, R., Hines, D. C., Lawrence, P. J., Terrile, R.J., Thomson, R. I. and Rieke, M. 1999, NICMOS Imaging of the HR 4796A Circumstellar Disk. Ap. J. 513, L1217-L130.

Tramontano, A. (2002) Private communication.

Tudge, C. (1991) Global Ecology, London: Natural History Museum Publications. p. 67.

Vogel, G. (1998) For island lizards, history repeats itself. Science, 279, 2043.

Yokoyama, R., and Yokoyama, S. (1990) Convergent evolution of the red- and green-like visual pigment genes in fish, Astynax fasciatus, and human. Proc. Natl. Acad. Sci. USA, 87, 9315-9318.