The most fundamental distinction in biology is between nucleic acids, with their role as carriers of information, and proteins, which generate the phenotype. In existing organisms, nucleic acids and proteins mutually presume one another. The former, owing to their template activity, store the heritable information: the latter, by enzymatic activity, read and express this information. It seems that neither can function without the other. Which came first, nucleic acids or proteins? There are three possible answers: (1) nucleic acids; (2) proteins; (3) neither: they coevolved. In this chapter, we discuss various possible answers to this 'chicken or egg?' problem. In section 5.2, we discuss what seems to us the most likely answer, that at first RNA performed both functions, as replicator and enzyme. In section 5.3, we consider an alternative view, in which protein enzymes existed either before, or alongside, the first nucleic acids. In section 5.4, we ask whether, perhaps, the first replicators were not nucleic acids. Finally, in section 5.5, we ask why, given that the genetic message is carried by nucleic acids, there are only four nucleotides and two base pairs. So far, we have tacitly assumed nucleic acids preceeded proteins, without stating the main reason. Nucleic acids came first because they can perform both functions: they are replicable, and they can have enzymatic activity. For many years, a common opinion was that to be replicable almost amounted to self-replicative ability, but that it was far-fetched to assume enzymatic activity. Today, there is increasing evidence that RNA can act as an enzyme, but we are more aware of the difficulty of self-replication. It should have been expected on theoretical grounds that RNA could act as an enzyme: the possibility was discussed by Woese (1967), Crick (1968) and Orgel (1968). Consider first why proteins can act as enzymes. An enzyme has a well-determined three-dimensional structure of chemical groups that, in most cases, arises automatically from the primary structure. Substrates of the enzyme are bound by the chemical groups on the surface. This means that the reactants will be kept in close proximity, and hence experience a much higher local concentration of each other than in solution. This by itself increases the rate of the reaction.