42 J. D. BERNAL 



nuclear membrane is also double. The association with DNA in all these cases 

 may be significant, indicating the relatively later arrival of this form of nucleic 

 acid. The disappearance of the nuclear membrane at the outset of mitosis may 

 also be significant as a reversion to an earher state of affairs. That most DNA- 

 containing structures are found v^^ithin such membranes may also be an indica- 

 tion of the relative lateness of DNA over RNA in biopoesis. 



One consequence of closed vesicles whose membranes offer a differential 

 resistance to the diffusion of molecules of different size or character is the 

 appearance of osmotic pressure and the need to equalize it by ionic mechanisms 

 or periodic expulsions of Uquid to secure any permanence for the enclosed 

 system. The simple membrane-covered drop has a minimimi surface for its 

 volimie. Where active molecules, such as enzjmies, are attached to it greater 

 effective surface can be obtained by pushing in the surface to form invaginations 

 or cisternae leading to the kind of complexity found in mitochondria, plastids, 

 Golgi apparatus, or endoplastic reticulum. At some such degree of complexity 

 and interdependence of biochemically specialized parts we approach the first 

 possibihty of a self-subsistent enclosed cell, an organism rather than an organelle. 

 How far the evolution was one single process, and how far it represents a fusion 

 of partially competent but chemically interdependent sub-organisms still remains 

 as a major problem in biopoesis. I am inclined to favour the latter alternative 

 and even to postulate something like an organization of this fusion through the 

 dominance of the more reliably self-reproducing DNA-protein mechanism. It 

 would seem from the study of plant viruses that RNA-protein reproduction, 

 precise as it is, can only take place in the presence of DNA and we as yet know of 

 no independent organisms which only contain RNA. 



It is perhaps best to leave the problem of physico-chemical biopoesis at this 

 point and refer the study of further development to the older estabhshed disci- 

 plines of morphological evolution. Even the development of the nucleated cell 

 of the protozoa and all higher organisms does not seem to depend on any radically 

 new biochemical factor. The specialization of parts of cells and ultimately of 

 whole cells for secretory, muscular, receptor or neural functions, seems to have 

 been achieved by the setting apart and modifying of the chemical and structural 

 elements already present in the generaHzed cell. 



The general picture of biopoesis described above seems to fall into seven 

 distinguishable stages set out in Table i. This provisional division is based 

 essentially on the degree of coherence and organisation of the effective sub- vital 

 or vital unit. In the first stage there is no degree of coherence. The whole or 

 part of the hydrosphere — oceans or lakes — is the seat of the essential micro- 

 molecular prevital process. In the second stage further biochemical evolution 

 occurs in concentrations determined by more or less fixed mineral aggregates 

 forming extensive sub-vital units. In the third stage, thanks to polymerization, 

 coacervate drops are formed Uberating life from mineral dependence and con- 

 stituting separate but indefinitely divisible and fusible eobionts. The fourth stage, 

 which I admit is more hypothetical than the others, is that of nucleoprotein 

 organelles similar to the viruses incapable of independent metabolism but set 

 apart from the remainder of the imiform protobiont. The fifth and later stages 



i 



