121 



atoms, S in proteins and P in the nucleic acids, are full partners of 

 the more usual atoms of organic molecules. The most abundant 

 mineral compounds of the planetary surface contain as well, atoms 

 still more abundant than are P and S, but not found at all within the 

 great polymers. Among these are in particular magnesium, silicon, 

 iron, and the sources of the ions of natural waters, especially Na and 

 K. These inorganic elements are involved in life today in a more or 

 less essential way, though they are not part either of protein or of 

 nucleic acid. The inorganic geochemists take these atoms as central 

 to their studies just as the biochemists look on the others as domi- 

 nant. Their possible interaction certainly should not be overlooked 

 simply because it does not quite lie within the most-studied disci- 

 plines of carbon chemistry. 



As pointed out in chapter V, the reduction both of carbon diox- 

 ide and nitrogen are possible on a primitive Earth, once the right 

 electron donors (e.g., Fe ) are present. The involvement of light to 

 drive the reactions is at least reminiscent of photosynthesis. These 

 photosynthetic models should be further explored. 



Replication 



As we have previously defined, a minimal living system capable 

 of evolution must be self-duplicating and mutable; it must have, at 

 least latently, the capacity for heterocatalysis. An example of such a 

 minimal definition has recently been met by an experiment in which 

 surface charge patterns on clay particles in water were shown to 

 replicate and mutate. The preliminary results of these experiments 

 indicate that clay minerals such as montmorillonite, which can swell 

 to a large degree, may be capable of replicative self-multiplication. 

 These minerals may, therefore, be looked upon as models for proto- 

 life, or possibly for most primitive life; their catalytic capabilities and 

 selectivities can be altered, and thus can accelerate or retard the rate 

 of self-multiplication. 



Results like those were foreseen by Cairns-Smith in 1965. He 

 suggested that the primitive genes were patterns of substitutions in 

 colloidal clay crystallites. The theoretical information density in 

 such crystallites is comparable to that in DNA. Evolution proceeded 

 through selective elaboration of pattern mutations that had survival 

 value for the clay crystallites that held them. 



