The important point is that hot and dry conditions were surely existing somewhere on the primitive Earth, for example in ponds and lakes that the Sun or volcanoes had dried up. It is reasonable, therefore, to expect that proteinoids did exist on our planet. But Fox went further, and proved that proteinoids can easily generate higher structures.
If a concentrated proteinoid solution is heated at between 120 and 200 degrees (centigrade) and then is very slowly cooled down, one can observe that proteinoids spontaneously form vesicles which Fox called microspheres. These structures come in fairly regular forms and dimensions (their diameters vary between 1 and 2 microns only), are very stable, and retain the bland catalytic activity of individual proteinoids. Despite the fact that lipids are absent, furthermore, a high number of vesicles exhibit semipermeable boundaries that look strangely similar to the lipid bilayer of true plasmatic membranes. But perhaps the most interesting thing is that microspheres can absorb proteinoids from the surrounding solution, which allows them to grow and eventually to divide in two by fission or budding (Figure 5-5).
Fox’s microspheres, in conclusion, are the first systems obtained in vitro that present a rudimentary type of metabolism. The evolutionary potential of the microspheres, however, remains a mystery. It is practically certain that they appeared on the primitive Earth, but we cannot be sure that they had a future, and it is for this reason that other solutions have been explored.
One of the most interesting is the theory of surface metabolism, an approach that was proposed, in different forms, by John Bernal in 1951, by Graham Cairns-Smith in 1982 and by Günther Wächtershäuser in 1998. The central idea of this theory is based on solid thermodynamic arguments. The formation of a peptide bond is not favoured in solution because it increases the entropy of the system, but, on a surface, the same process takes place with a decrease of entropy, and is therefore favoured. And this is true not only for peptide bonding but for many other types of polymerization. A great number of enzymatic reactions require a collision of three molecules, an event which is highly unlikely in space but much more probable on a surface.