Tuning into the_en frequencies of life: does life on Earth have cosmic implications?

by Simon CONWAY MORRIS, Professor of Palaeontology at the University of Cambridge, UK

An important text by one of the leading contemporary specialists of evolution who claims that if we replay evolution it would occur in a similar way and lead to consciousness.

Abstract. While the fitness of the environment has clear implications for the molecular basis of life and the resultant biochemistry, its extrapolation to complex systems that characterize all organisms appears to be far more tenuous. Such organisms may be adapted to their local environment, but it is widely agreed that they are effectively fortuitous end-products of a process with an almost indefinite number of trajectories. I disagree, and here I attempt to explore the connections between the molecular substrate and evolutionary convergence, that is how the universal meets the inevitable.

Introduction Philosophers like to joke that much of what they profess to do is little more than scribble footnotes to Plato. So too, in the eyes of some, biology has a similar status with respect to the physical sciences. Thus the fundamentals of the material world are codified in the immutable laws of physics and the Periodic Table, but once one moves towards biological systems then matters become much more indeterminate. To be sure living organisms must be governed by the inexorable realities of the physical world, such as Reynolds numbers or the principles of fluid flow in pipes, but so far as the end-results – or at least the three billion year experiment of which we are an infinitesimally small part – are concerned, then there is no predictability. Humans, sperm-whales and tulips are all evolutionary accidents.

That, at least, is the prevailing paradigm. It is one which emphatically rejects a fine-grained view of evolution. Whilst there are certainly traditions that regard the evolutionary process as almost entirely open-ended, a position tenaciously adopted by S.J. Gould, a more realistic assessment is content to identify some broad constraining boundaries. Thus at least some anthropic principles proclaim that the Universe is “just right” for life. Even if life itself is confined to planetary surfaces, it is also widely regarded as a cosmic imperative (e.g. de Duve 1995). In a slightly more local chemical context Williams and Fraústo da Silva (2003) have argued that thermodynamics and the rules of chemical assembly impart a strong directionality to evolution. From the point of view of the organism the principal constraints revolve around the nature of intracellular reductive chemistry, the challenge of oxygen, and the advantages of co-operative interactions in the context of ecosystems. Together, Williams and Fraústo da Silva (2003) argue, they led to “changes that were in an inevitable progression, and were not just due to blind chance, (p. 323), with the implication that even to a background of mutational or other Darwinian change there was still “a fixed overall route”. In this way they suggest that not only was the emergence of eukaryotes inevitable, but so also were plants and animals. While it seems to be implicit in their analysis that humans are a logical outcome of this process, nowhere do they state that such an evolutionary outcome is inevitable. Thus while this paper is exceptionally important in its claim that “overall activity of the blind watchmaker was constrained by the nature of changing chemicals and the thermodynamic equilibrium conditions of the environment” so that “Life was in a physical tunnel and there was only one way to go” (Williams and Fraústo da Silva 2003, p. 335), it does not make claims for particular organismal specificities. Nor, I hasten to add, would I necessarily expect Williams and Fraústo da Silva to address this topic : their arguments are effectively chemical and thermodynamic.

Nevertheless, Williams and Fraústo da Silva’s (2003) analysis is very much in the spirit of Lawrence Henderson’s (1913) admiration for the way the physico-chemical glove matches the hand of life. In this review I want to address this metaphor, and see how far the specificities of biological organizations are inevitable products of the evolutionary process. This is an issue reviewed at some length elsewhere (Conway Morris 2003), and rather than repeat some of these conclusions, I want to explore some issues I mostly dealt with rather cursorily in Life’s Solution, or not at all because new information has only just become available.

Some basic assumptions

While I am very intrigued by the direction in which this topic appears to be moving, there are a number of presuppositions whose significance is difficult to assess. These revolve around questions of predeterminants, pathways, incumbencies, and inevitabilities. From a historical perspective, evolution is faced with repeated “decisions”, and it is not clear to what extent a pre-existing substrate, say a particular protein family, is absolutely necessary for the ultimate emergence of a given complexity, say intelligence. Nor is it clear whether evolutionary incumbency, that is the pre-occupation of a metaphorical “high ground”, can in principle permanently frustrate the emergence of a particular complex system. Nevertheless, the evidence (Conway Morris 2003) seems very much to point in the other direction, that is, towards inevitabilities of outcome. This conclusion is based on the observation of evolutionary convergences, but I suspect the more interesting conclusion is that “navigation” by evolution is predetermined by much deeper structures that effectively define a “road-map” for life.

What then might be the basis for such a “road-map” ? I will start with the assumption that all life is carbaquist, thus dependant on the remarkable properties of water (Henderson 1913) and the ability of carbon to form both flexible chains and chemical bonds. Liquid ethane and silicon are amongst the alternatives, but appear to suffer significant drawbacks. Concerning those regions of the “road-map” of which we know the least, that involving the origin of life might as well be labeled “Here be dragons”. My hunch is that as and when this pathway is finally elucidated, it will turn out to be extraordinarily specific, involving the by now customary happy “coincidences”. At present, however, the diversity of experimental approaches are clear indications that both the process of assembly and place of assembly are speculative.

The difficulties in understanding the origin of life are all the more puzzling because there is some evidence that the basic parameters are probably universal. In part this is because of the ease of synthesis of many of the principal building blocks of life, although in the context of the origin of life “industry” all too often the difficulties of very low yields, unwanted by-products, and widely different stability fields of many key molecules, are too easily brushed aside. Nevertheless, the experimental production and detection in carbonaceous meteorites of amino acids, nucleic acids (or precursors) and hydrocarbons, suggest that this is the molecular substrate upon which all life sits. So far as the amino acids are concerned, one estimate is that approximately three-quarters of those found in terrestrial life would have extra-terrestrial counterparts (Weber and Miller 1981). So too in his research programme of looking at the so-called etiology of DNA the team led by Albert Eschenmoser (1999) shows that few of the alternatives match the effectiveness of DNA itself. It is worth remembering three implications of this research programme. First, as Eschenmoser (1999) stresses, the principal of molecular choice revolves around optimization, rather than maximization, of such properties as conformational flexibility. Second, even if there are alternative DNAs that match, or even exceed, the optimized properties of “real” DNA this does not necessarily mean this advantage extends to the operational milieu of the chromosomes and the cell. Finally, while various alternatives to the sugar and nucleic acid building blocks are also feasible, in at least some cases their synthesis in any pre-biotic situation seems improbable. This is not to say, of course, that DNA was not preceded by a viable and more primitive precursor, such as TNA (Schöning et al. 2000), but here too questions of possible chemical pathways and ease of synthesis cannot be ignored.

Universal biochemistry ?

What of the next stage, the integration of these molecules into a functioning biochemistry ? Here Pace (2001) has made a strong argument for its universality. Such a view is probably relatively uncontroversial, and echoes Wald’s famous remark that candidates for biochemistry exams would be at equal advantage whether they sat them on Earth or Arcturus. In a somewhat different vein Denton et al. (2002) have argued that the basic protein folds are an axiomatic product of amino-acid assembly, constrained by constructional rules that impart a law-like behaviour that reflects innate, and inevitable, tendencies to particular types of organization. The extent to which such constraints apply also to specific types of protein, such as seven-helical, membrane-spanning or globins, which are often associated with particular functions (in these two cases respectively signal transduction and oxygen transport) remains to be seen. The evidence for functional convergences in these and other proteins (Conway Morris 2003) suggests that here too there may be important rules of assembly.

While such universalities may not be supported by all biologists, the idea of the molecular architecture of alien biospheres being similar, perhaps very similar, to that of Earth would not come as a complete surprise. Thereafter, however, the likelihood of precise correspondences between the biospheres of these two worlds would be regarded as increasingly improbable. To illustrate this consider the following list of connected examples : photosynthesis, chlorophyll, chloroplasts, water conducting tissue (xylem), flowers and a rose garden. Most evolutionary biologists would, I suspect, see this list as one of ever-decreasing evolutionary probabilities. So far as planetary life is concerned, photosynthesis may well be universal (see also Wolstencroft and Raven 2002). Such may also be true of chlorophyll (Wald 1974), and although proteorhodopsin can also as a phototrophic molecule (Beja et al. 2001) its use is restricted to particular bacteria and was co-opted from its prior use as a proton pump. Moreover, while it may well provide an energetic advantage to the bacteria, there is apparently no evidence that it can actually assist in the fixation of carbon. Beyond photosynthesis, and possibly chlorophyll, the remainder of my list would be regarded as a series of fortuitous evolutionary innovations, of only terrestrial significance. Such a view might be premature. Thus, although generally regarded as monophyletic, Stiller et al. (2003) present evidence that chloroplasts may have arisen independently several times. So too xylem has evolved twice (Ligrane et al. 2002), as have flowers (see Conway Morris 2003, pp. 135-138). On this basis only the rose garden is unique, although here we need to remind ourselves that flowers also show recurrent homoplasies (e.g. Hufford 1997).

The extent to which the fitness of the environment matches the fine-grained nature of life is an area about which we know far too little. Discussion is hindered not only because of the inevitable specializations in biology – watching a biochemist, ecologist and palaeontologist in animated conversation is an all-too-rare sight – but also the uncertainty of what depends on what. Thus, the gene-centred view, made popular by Dawkins, is not only over-reductionist, but in some cases is potentially seriously misleading. Consider, for example, the classic case of haplodiploidy as an explanation of the highly organized colonies seen in eusocial insects. Despite the elegance of this hypothesis and its neat explanation of how non-reproductive individuals retain a genetic benefit even though they cannot pass on directly their genes, an alternative explanation for kin selection in terms of a benefit conferred by a balance between individual risk and colony viability (a sort of life insurance) is, at least in some circumstances, a more convincing explanation for the origin and maintenance of eusociality (e.g. Queller and Strassman 1998 ; Field et al. 2000 ; Landi et al. 2003). The point here is not to dispute the reality of kin selection, but to stress that the life insurance hypothesis, together with other features such as worker policing and individuality (in terms of recognition and apparent choice of activity) remove the reductionist stamp to eusociality with the implicit assumption of blind and robotic forces. Alternative views also seem, however, to be incomplete. For example, the debate of “molecules versus morphology” remains inconclusive, despite the remarkable advances in developmental biology. The truth surely is that our growing understanding of adaptive complexes only serves to reinforce our (or at least my) wonder at the robustness, integration and sophistication of biological systems ; no wonder they are so hard to dissect. Moreover, while the adaptive explanation (and penalty for failure) in these systems is self-evident, it is worth reminding ourselves of the immense difficulties that confront any ab initio imitation of these systems, perhaps most obviously in the arena of artificial intelligence.

Molecular convergence

In an attempt to address how the fitness of the environment applies not only to biochemistry, but has implications for the inevitable emergence of complex systems, I will now address several relevant topics. The first concerns the rather remarkable evidence, only now emerging, for various sorts of molecular convergence. To the first approximation this would be expected to be very unusual indeed, for the simple reason that with the number of alternatives being astronomically large, the coincidence of even a short identical sequence is very small indeed. In reality there are considerably more examples than might be thought (see Conway Morris 2003), although this is much less surprising given functional constraints, such as those associated with active enzymatic sites (for a recent example see Beuth et al. 2003). A particularly interesting case concerns an example of convergent evolution in a virus, specifically the bacteriophage ?X174 (Bull et al. 1997). In one sense this example is somewhat artificial in as much as it is laboratory based, and is designed to study the evolutionary response to the change in a rather general environmental property, in this case an increase in temperature. Study of five lineages associated with two bacterial hosts (E. coli, Salmonella typhimurium) showed striking convergence at a few key sites in the genome. The response to heat, therefore, was evidently selective. Given the widely acknowledged (although remarkably unexplored, but see Axe 2000) reality of extreme sensitivities of certain molecular sites, such convergences are actually unsurprising (even though the scientist’s reaction is almost always precisely the opposite, Conway Morris 2003, p. 128). As Bull et al. (1997) also point out, despite the generality of the imposed environmental change, the experiments are strongly constrained by both the similar selective environment (specifically an increase from 38°C to 43.5°C) and the fact that the starting point of each viral lineage had a practically identical genome. On the other hand it is important to note that without the prior knowledge of the convergence the true evolutionary tree of divergence would have been unrecoverable. Moreover, whilst Bull et al. (1997) stress the unusual nature of this case, they also remark that molecular convergences may be more prevalent than generally thought.

Evidence for molecular convergence is also emerging in other quarters, including both transcriptional machinery and developmental processes. In the context of the former particularly important is the analysis by Conant and Wagner (2003) of gene circuits in E. coli and S. cerevisiae (yeast). Of the four circuit types, e.g. bi-fan, feed-forward in yeast (and two equivalents in E. coli) the evidence strongly points to the great majority having independent origins. As Conant and Wagner (2003) remark “Our results also suggest that convergent evolution … may have an important role in the higher organizational level of gene circuits. Stephen Jay Gould famously asked what would be conserved if life’s tape, its evolutionary history, was replayed…. Transcriptional regulation circuits, it seems, might come out just about the same” (p. 265).

Whilst these examples of molecular convergence are ultimately less surprising than might be at first thought, the emerging evidence for various sorts of convergence in developmental biology is attracting considerable attention, principally because of the continuing discussions of the nature of biological homology and the roles of constraint. Broadly the evidence for possible convergence in developmental mechanisms falls into three categories. The first concerns the repeated and independent emergence of particular phenotype, but on the basis of the same regulatory mechanism. Examples may be quite specific, for instance, the loss of trichomes in a larval fly (Sucena et al. 2003), or more general as in the evolution of lecithotropic, direct-developing sea-urchin larvae (Wray 2002 ; see also Nielsen et al. 2003). The focus of interest here is really the reiteration of a phenotype, presumably in an adaptive context, and the unsurprising observation that a particular developmental pathway is recruited “on demand”. The homoplasic spanner this throws into the painfully literal cladistic machinery will be self-evident. Whilst these convergences involve animals, a comparable result has also been documented in a microbial context. This involved a study of directed evolution (for 20,000 generations) in twelve populations of E. coli, where rather remarkably 59 genes showed parallel changes in expression patterns, and all in the same direction (Cooper et al. 2003).

The second category is potentially more significant, because here a very similar phenotype emerges from a different developmental basis (e.g. Gompel and Carroll 2003 ; Wittkopp et al. 2003). This, of course, is central to convergence of phenotypes where the route navigated is very different (at some level), but the “solution” is much the same (Conway Morris 2003). The third category, which is more an extension of the first instance, is where a complex structure is “built” by the recruitment of similar genetic modules. The almost identical construction of the insect wing and vertebrate limb might be one such example (see Tabin et al. 1999 for a discussion of the various alternatives).

Closely linked to the question of convergence (and constraint) in developmental systems is clear evidence for repeated co-option of developmental genes for new functions. So widespread is this phenomenon that it raises major difficulties in determining the “primitive” function of a given gene. As importantly, it is difficult to decide what, if any, are the general rules of engagement. In some specific cases, and as already discussed in the context of convergence, it looks as if the organism in question had “no choice”. This may be the case in the repeated recruitment of the gene otx to tube-foot development in direct developing sea-urchins (Nielsen et al. 2003). At present other examples of apparent redeployment, such as genes essential for the development of the eye being recruited for the expression of spermatocyte (Fabrizio et al. 2003) and muscle (Heanue et al. 1999 ; see also Relaix and Buckingham 1999) tissue, are presumably fortuitous. Alternatively, there may be much deeper constraints that make the co-option of these genes (and their protein products) effectively inevitable.

Evolutionary inherency

Much of the preceding discussion hinges crucially on the question of when in evolutionary history a particular molecule or developmental system appeared. In the case of convergent gene circuits (Conant and Wagner 2003) the possibility that there were a few ancestral circuits that diversified by gene duplication can be rejected. In the majority of cases, however, it is still unresolved how deep in evolutionary history a character can be traced, and what key steps were required (and in what order) for its involvement in the emergence of a complex structure. Despite this relative uncertainty, it is clear that a number of molecules and molecular mechanisms that are essential in the success of advanced groups evolved at an earlier stage and have been elaborated, often in distinct ways, in different lineages. Such is the case, for example, with glumate-based cell signaling (Dennison and Spalding 2000) and the repeat protein known as armadillo (Coates 2003). In this way a great deal of the complexity of animals and plants is evidently inherent at a microbial level. Thus while animals are largely defined by such properties as a nervous system, sensory organs and contractile tissue such key molecules as acetylcholine (Wessler et al. 1999), crystallins (e.g. Piatigorsky 1992) and myosin (Berg et al. 2001) had evolved hundreds of millions, if not billions, of years before animals themselves.

Social microbes, intelligent plants ?

The emergence of these complex systems has another interesting dimension in the form of the evolution of analogous systems. In one way these similarities are hardly surprising, given that they call upon a common molecular repertoire. They remain important, however, because they give us some sense of the range of alternatives and a healthy reminder that while animals may be the only group with the potential to understand Creation the other players reinforce our sense of its richness, diversity and even strangeness. Thus Crespi (2001 ; see also Rainey and Rainey 2003 ; Velicer and Yu 2003) notes how it is that “all the hallmarks of a complex and coordinated social life” (p. 178) are identifiable in microbial communities, with analogues of cooperation, division of labour, communication and sociality being identifiable. Crespi (2001) identifies seven social phenomena that represent behavioural convergences with “higher” organisms, and as importantly suggests that these are indicative of adaptation.

In an even more ambitious vein Trewavas (2003) explores how aspects of communication, computation and intentionality in plants reflect not only their adaptive plasticity but also a type of intelligence. At first sight his analysis looks to be decidedly heterodox, but his fundamental point is that the many analogues of animal intelligence, such as stimulus response and memory, have a similar chemical basis. The analogues are just that, and Trewavas (2003) offers not only some interesting speculations on the future understanding of plant communication, but also the way in which the networks offered by separate ramets allow the potential for a sort of parallel processing.

Evolutionary inevitabilities

These inherencies and analogues suggest that the motors of adaptation and ecological diversification make the emergence of complex biological systems, say an eye seeing a rose garden, very probable, perhaps inevitable. The basic similarity of these analogues indicates that radical, alien alternatives may be much less likely than is often thought. Such arguments would, I suggest, apply to the emergence of all biological complexities, including intelligence (Conway Morris 2003) and as Grammer et al. (2003) have recently argued also the appreciation of beauty. Concerning the evolutionary inevitability of intelligence there now appears to be specific support from experiments designed to establish whether cognitive functions are uniquely propositional and so unable to function in the absence of symbolic-based language with implicit semantics and syntax, or whether they are fundamentally an analogue-coding process whereby there is a continuum between cognitive abilities and lower level sensory mechanisms. One way to test this is to examine numerosity, i.e. the ability to count, in animals because this property presumably involves both the sensory perception (“I see two dogs”) and abstraction (“I see two extraterrestrials, not two dogs”). Such experiments (e.g. Nieder and Miller 2003) indicate the analogue-coding hypothesis is correct. At the moment such results make the evolutionary emergence of human cognition unremarkable, but also fail to explain why in certain important respects it is unique.

Future work

In the context of this article there are a number of other areas that I believe also hold some promise. One already alluded to briefly, concerns the question of molecular sensitivities whereby a trivial change, sometimes the substitution of a single amino acid in a protein, can lead to its catastrophic failure. So far as I am aware, this has not been investigated in a systematic fashion, but such sensitivities are apparently at odds with the evidence of extensive substitutions, sometimes at a sufficiently regular rate to allow the use of molecular “clocks”. The standard answer is that the “scaffolding” of a protein is almost entirely mutable so long as it maintains its structure, whereas the active biochemical sites are necessarily highly constrained. The results of Axe (2000), however, should give us pause for thought, and serve to re-open the question of the functionality and integration of the entire protein molecule.

Another intriguing question is whether the fitness of the environment confers any predictability to the evolutionary process. Thus, the likelihood of a universal biochemistry and the inevitability of organismic convergence at many levels (Conway Morris 2003) may provide points for discussion. In an interesting analysis Hofman (2001) has argued that the potential for further encephalization in the hominid is limited. The stock response, of course, is to postulate computer-based extensions of our cognitive ability. A more radical alternative is to suggest that in the evolution of (any) biosphere the rise of bird – and mammal-like intelligences with a potential for technologies is geologically short-lived and is replaced by other systems, such as insect eusocial colonies.

Coda

In conclusion, I wish simply to comment that while the fitness to the environment is most obvious and direct at the level of biochemistry, the interconnections of life via the evolutionary process effectively pre-determine the process and thereby impart both directionality and inevitability. In brief, wherever there is life, there will, in due course, be mind. Whether it is always our mind is another question.
REFERENCES

Axe, D.D. 2000. Extreme functional sensitivity to conservative amino acid changes on enzyme exteriors. Journal of Molecular Biology 301 : 585-596.

Beja, O., Spudich, E.N., Spudich, J.L., Leclerc, M. and DeLong, E.F. 2001. Proteorhodopsin phototrophy in the ocean. Nature 411 : 786-789.

Berg, J.S., Powell, B.C. and Cheney, R.E. 2001. A millennial myosin census. Molecular Biology of the Cell 12 : 780-794.

Beuth, B., Niefind, K. and Schomburg, D. 2003. Crystal structure of creatinase from Pseudomonas putida : A novel fold and a case of convergent evolution. Journal of Molecular Biology 332 : 287-301.

Bull, J.J., Badgett, M.R., Wichman, H.A., Huelsenbeck, J.P., Hillis, D.M., Gulati, A., Ho, C. and Molineux, I.J. 1997. Exceptional convergent evolution in a virus. Genetics 147 : 1497-1507.

Coates, J.C. 2003. Armadillo repeat proteins : beyond the animal kingdom. Trends in Cell Biology 13 : 463-471.

Conant, G.C. and Wagner, A. 2003. Convergent evolution of gene circuits. Nature Genetics 34 : 264-266.

Conway Morris, S. 2003. Life’s solution : inevitable humans in a lonely universe. Cambridge University Press.

Cooper, T.F., Rozen, D.E. and Lenski, R.E. 2003. Parallel changes in gene expression after 20,000 generations of evolution in Escherichia coli. Proceedings of the National Academy of Sciences, USA 100 : 1072-1077.

Crespi, B.J. 2001. The evolution of social behavior in microorganisms. Trends in Ecology & Evolution 16 : 178-183.

Dennison, K.L. and Spalding, E.P. 2000. Glutamate gated Ca2-1 fluxes in Arabidopsis. Plant Physiology 124 : 1511-1514.

Denton, M.J., Marshall, C.J. and Legge, M. 2002. The protein folds as Platonic forms : New support for the pre-Darwinian conception of evolution by natural law. Journal of Theoretical Biology 219 : 325-342.

de Duve, C. 1995. Vital dust : Life as a cosmic imperative. BasicBooks HarperCollins : New York.

Eschenmoser, A. 1999. Chemical etiology of nucleic acid structure. Science 284 : 2118-2124.

Fabrizio, J.J., Boyle, M. and DiNardo, S. 2003. A somatic role for eyes absent (eya) and sine oculis (so) in Drosophila spermatocyte development. Developmental Biology 258 : 117-128.

Field, J., Shreeves, G., Sumner, S. and Casiraghi, M. 2000. Insurance-based advantage to helpers in a tropical hover wasp. Nature 404 : 869-871.

Gompel, N. and Carroll, S.B. 2003. Genetic mechanisms and constraints governing the evolution of correlated traits in drosophilid flies. Nature 424 : 931-935.

Grammer, K., Fink, B., Møller, A.P. and Thornhill, R. 2003. Darwinian aesthetics : sexual selection and the biology of beauty. Biological Reviews 78 : 385-407.

Heanue, T.A., Reshef, R., Davis, R.J., Mardon, G., Oliver, G., Tomarev, S., Lassar, A.B. and Tabin, C.J. 1999. Synergistic regulation of vertebrate muscle development by Duch2, Eya2, and Six1, homologs of genes required for Drosophila eye formation. Genes and Development 13 : 3231-3243.

Henderson, L.J. 1913. The fitness of the environment : An inquiry into the biological significance of the properties of matter. Macmillan.

Hofman, M.A. 2001. Brain evolution in hominids : are we at the end of the road ? In Evolutionary anatomy of the primate cerebral cortex (D. Falk and K.R. Gibson, eds), pp. 113-127. Cambridge University Press.

Hufford, L. 1997. The roles of ontogenetic evolution in the origins of floral homoplasies. International Journal of Plant Sciences 158 (Suppl. 6) : 565-580.

Landi, M., Queller, D.C., Turillazzi, S. and Strassman, J.E. 2003. Low relatedness and frequent queen turnover in the stenogastrine wasp Eastenogaster fraterna favor the life insurance over the haplodiploid hypothesis for the origin of eusociality. Insectes Sociaux 50 : 262-267.

Ligrane, R., Vaughn, K.C., Renzaglia, K.S., Knox, J.P. and Duckett, J.C. 2002. Diversity in the distribution of polysaccharide and glycoprotein epitopes in the cell walls of bryophytes : new evidence for the multiple evolution of water-conducting cells. New Phytologist 156 : 491-508.

Nieder, A. and Miller, E.K. 2003. Coding of cognitive magnitude : Compressed scaling of numerical information in the primate prefrontal cortex. Neuron 37 : 149-157.

Nielsen, M.G., Popodi, E., Minsuk, S. and Raff, R.A. 2003. Evolutionary convergence in otx expression in the pentameral adult rudiment in direct-developing sea urchins. Development, Genes and Evolution 213 : 73-82.

Pace, N.R. 2001. The universal nature of biochemistry. Proceedings of the National Academy of Sciences, USA 98 : 805-808.

Piatgorsky, J. 1992. Lens crystallins. Innovation associated with changes in gene regulation. Journal of Biological Chemistry 267 : 4277-4280.

Queller, D.C. and Strassman, J.E. 1998. Kin selection and social insects. BioScience 48 : 165-175.

Rainey, P.B. and Rainey, K. 2003. Evolution of cooperation and conflict in experimental bacterial populations. Nature 425 : 72-74.

Relaix, F. and Buckingham, M. 1999. From insect eye to vertebrate muscle : redeployment of a regulatory network. Genes and Development 13 : 3171-3178.

Schöning, K-U., Scholz, P., Guntha, S., Wu, X., Krishamurthy, R. And Eschenmoser, A. 2000. Chemical etiology of nucleic acid structure : the ?-threofuranosyl-(3 ?-2 ?) digonucleotide system. Science 290 : 1347-1351.

Stiller, J.W., Rees, D.C. and Johnson, J.C. 2003. A single origin of plastids revisited : Convergent evolution in organellar genome content. Journal of Phycology 39 : 95-105.

Sucena, E., Delon, I., Jones, I., Payre, F. and Stern, D.L. 2003. Regulatory evolution of shavenbaby/ovo underlies multiple cases of morphological parallelism. Nature 424 : 935-938.

Tabin, C.J., Carroll, S.B. and Panganiban, G. 1999. Out on a limb : Parallels in vertebrate and invertebrate limb patterning and the origin of appendages. American Zoologist 39 : 650-663.

Trewavas, A. 2003. Aspects of plant intelligence. Annals of Botany 92 : 1-20.

Velicer, G.J. and Yu, Y-t.N. 2003. Evolution and novel cooperative swarming in the bacterium Myxococcus xanthus. Nature 425 : 75-78.

Wald, G. 1974. Fitness in the Universe : Choices and necessities. Origins of Life and Evolution of the Biosphere 5 : 7-27.

Weber, A.L. and Miller, S.L. 1981. Reasons for the occurrence of the twenty coded protein amino acids. Journal of Molecular Evolution 17 : 273-284.

Wessler, I., Kirkpatrick, C.J. and Racke, K. 1999. The cholinergic ’pitfall’ : Acetylcholine, a universal cell molecule in biological systems, including humans. Clinical and Experimental Pharmacology and Physiology 26 : 198-205.

Williams, R.J.P. and Fraústo da Silva, J.J.R. 2003. Evolution was chemically constrained. Journal of Theoretical Biology 220 : 323-343.

Wittkopp, P.J., Williams, B.L., Seleque, J.E. and Carroll, S.B. 2003. Drosophila pigmentation evolution : Divergent genotypes underlying convergent phenotypes. Proceedings of the National Academy of Sciences, USA 100 : 1808-1813.

Wolstencroft, R.D. and Raven, J.A. 2002. Photosynthesis : Likelihood of occurrence and possibility of detection on Earth-like planets. Icarus 157 : 535-548.

Wray, G.A. 2002. Do convergent developmental mechanisms underlie convergent phenotypes ? Brain, Behavior and Evolution 59 : 327-336.