The difficulties suggested by J. Von Neumann’s analysis will be taken as an indication that the replication of primitive organisms was largely or wholly imposed on them by their environment — that the process of replication per se did not require the pre-existence of any elaborate internal instructions or machinery within the organisms. It has been argued that clay crystallites may have constituted the genes of primitive organisms, with survival-promoting instructions written in the form of patterns of internal cation substitutions, and with phenotypes consisting of adsorbed, abiogenically formed, organic molecules. Genetic extension could increase the versatility of a primitive genetic material, but there is no particular reason to suppose that such a material — one that is a good starter — would be particularly good at evolving very far in this way. The ‘metamorphosis’ occurred through an extended evolution during which the original organisms became heterogenetic.